[go: up one dir, main page]

CN102278995A - Bayes path planning device and method based on GPS (Global Positioning System) detection - Google Patents

Bayes path planning device and method based on GPS (Global Positioning System) detection Download PDF

Info

Publication number
CN102278995A
CN102278995A CN2011101066426A CN201110106642A CN102278995A CN 102278995 A CN102278995 A CN 102278995A CN 2011101066426 A CN2011101066426 A CN 2011101066426A CN 201110106642 A CN201110106642 A CN 201110106642A CN 102278995 A CN102278995 A CN 102278995A
Authority
CN
China
Prior art keywords
path
road
gps
good
bayesian
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
CN2011101066426A
Other languages
Chinese (zh)
Other versions
CN102278995B (en
Inventor
曾喆
李清泉
万剑华
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.)
China University of Petroleum East China
Original Assignee
China University of Petroleum East China
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 China University of Petroleum East China filed Critical China University of Petroleum East China
Priority to CN 201110106642 priority Critical patent/CN102278995B/en
Publication of CN102278995A publication Critical patent/CN102278995A/en
Application granted granted Critical
Publication of CN102278995B publication Critical patent/CN102278995B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Traffic Control Systems (AREA)

Abstract

本发明涉及一种基于GPS探测的贝叶斯路径规划装置及方法。本发明方法是采用先验知识推估后验信息的方法来模拟司机的一种先验找路方法；本装置利用贝叶斯分类器将司机的先验推断方法纳入到我们的路径规划装置中，其本身是基于分类器装置对路网做模式识别处理后而形成的路径规划方法。本发明利用这种司机对道路网络的实时交通状况的先验推断，构造了贝叶斯分类器来形成对道路网络的优先识别，基于这种优先识别而形成的一种路径规划装置及方法，它可以根据高峰期、非高峰期的先验知识在不同的交通时段中分别给出不同的最优路径。该装置及方法具有良好的实用性，并可以灵活的集成到各种基于道路交通的导航系统中。

Figure 201110106642

The invention relates to a Bayesian path planning device and method based on GPS detection. The method of the present invention is a method of simulating the driver's priori wayfinding by using priori knowledge to estimate the posteriori information; the device uses the Bayesian classifier to incorporate the driver's priori inference method into our path planning device , which itself is a path planning method formed after the classifier device performs pattern recognition processing on the road network. The present invention utilizes the driver's prior inference of the real-time traffic conditions of the road network, constructs a Bayesian classifier to form a priority recognition of the road network, and forms a path planning device and method based on the priority recognition, It can give different optimal routes in different traffic periods according to the prior knowledge of peak hours and non-peak hours. The device and method have good practicability and can be flexibly integrated into various navigation systems based on road traffic.

Figure 201110106642

Description

Bayes's path planning apparatus and method based on the GPS detection
Technical field:
The present invention relates to a kind of a kind of path planning apparatus and method of adopting based on the deduction of priori of being suitable under actual traffic trip situation, particularly a kind of Bayes's path planning apparatus and method of surveying based on GPS.
Background technology:
Current, GPS (Global Positioning System) should use as positioning navigating device on various vehicles widely, such as onboard navigation system, and the locating and monitoring system of vehicle, Floating Car system of traffic department or the like.On the other hand, the optimal path that obtains of the method for optimizing route in our the various navigational system its path of planning out a kind of shortest path on absolute time or length tolerance often in actual use.Yet because the random character of the real-time traffic of road, the optimal path on the tolerance meaning often is not the priori that meets the driver so like this, and such optimal path often is not the optimal path of our real world yet.
Patent (Chinese patent application number 200410103954.1) is that the time of vehicle and speed are merged a kind of method into path planning, patent (Chinese patent application number 200610036181.9) be a kind of be a kind of paths planning method that transit time is realized with the highway section weight setting, from top this several patents application form, it only is a kind of path optimization method that provides on tolerance (length or time) that some patents about path planning apparatus and method, existing patent about path planning have been arranged at present.These methods all do not have to adopt based on Bayesian priori deduction method comes realizing route planning.
Announced one piece document " taxi experimental knowledge modeling and path planning algorithm " in the volume (404-409 page or leaf) " mapping journal " Augusts 39 in 2010, though some has considered the posterior infromation of taxi, but only be some information to have been arrived under the road on the weights W with following computing formula reduction I(t)=[μ 1T I(t)+ μ 2S i]/μ 3C I (t)
In essence, this formula is that time and velocity information are passed through this formula, and the w weight setting to the highway section of correspondence, is calculated optimal path by traditional dijkstra method at last.The method of path planning being done in the posterior infromation reduction to the road weights, possessed under the situation of Real-time Traffic Information, is the same with the problem of traditional optimal path method, and it can't avoid the randomness of Real-time Traffic Information.This method is not a kind of priori deduction method, thereby and adopts the calculating of the optimal path planning on the tolerance to distinguish also little on method.Therefore, the document does not form a kind of driver's prior imformation and according to the Bayesian inference road collection of classifying, and carries out path planning in view of the above.
Technical method from the existing path planning of last surface analysis, these are with speed, length, the optimal path method of certain posterior infromation naturalization such as transit time on the tolerance of road weights, a kind of priori deduction method of the driver that really will be familiar with very much the road network traffic is not brought in the path planning system.
Summary of the invention:
Purpose of the present invention is exactly the above-mentioned defective that exists at prior art, a kind of Bayes's path planning apparatus and method of surveying based on GPS is provided, and the inventive method is that a kind of priori that the method that adopts priori to estimate posterior information is simulated the driver is looked for path method; This device utilizes Bayes classifier that driver's priori deduction method is brought in our path planning apparatus, itself is based on the sorter device road network is done after the pattern recognition process and the paths planning method that forms.
Bayes's path planning apparatus of surveying based on GPS of the present invention mainly is divided into two parts of end of probe and center processing end, and wherein end of probe is selected the driver that some are familiar with road traffic condition, the GPS detector is distributed on these drivers' the vehicle; Described center processing end comprises receiver, central control unit, path screening washer, Bayes classifier and the several modules of optimal path computation device:
End of probe wherein: the GPS detector mainly is to come n driver's driving trace is surveyed with the GPS collector;
Described center processing end: central control unit, the data of controlling between other the module transmit; The path screening washer, the GPS tracing point of these driver's vehicles that main basis collects and the starting point and the terminating point of collection recover the path of travelling, and filter out " good " path fast by speed, that expend time in short; Bayes classifier is practiced Bayes classifier according to " good " path training of setting up, and according to the Bayes classifier that trains road is done classification, forms preferential road collection, suboptimum road collection and whole road collection; The optimal path computation device according to three kinds of preferred roads that form, is implemented optimum route search and is calculated, and obtains the optimal path result; Path memory is deposited " good " path collection that the path screening washer is set up; The classification of road storer is deposited three kinds of road collection that Bayes classifier classification back forms.
Bayes's paths planning method of surveying based on GPS of the present invention may further comprise the steps:
(1), end of probe is obtained the trip empirical data with the GPS detector;
(2), the center control end obtains track data and the realizing route planning by handle a series ofly from the GPS detector, may further comprise the steps:
(2.1) obtain the path collection of " good " on the priori with the path screening washer: mainly recover the path of travelling, and filter out " good " path fast by speed, that expend time in short according to the GPS tracing point of these driver's vehicles that collect and the starting point and the terminating point of collection;
(2.2) practice Bayes classifier according to " good " path training of setting up, road is done classification, form preferential road collection, suboptimum road collection and whole road collection according to the Bayes classifier that trains;
(2.3) calculate optimal path in the optimal path computation device, central control unit is regarded the preferential road collection of above-mentioned three classes as the triple-decker network and is input to the optimal path computation device, obtains the optimal path result;
Above-mentioned path screening washer mainly is made up of comparison controller, coupling computing module and buffer three parts:
A) in the coupling computing module, the driver's who collects according to end of probe the GPS track and the starting point and the terminating point of record, with map-matching method it is matched the road corresponding highway section, and join according to adjacent road section and to obtain the path that forms between each starting point and the terminating point, the path collection that the coupling computing module obtains is input in the comparison counter;
B) in comparison controller, elder generation can obtain list according to the journey time of the total length and the GPS record of every paths
The average passage rate v that paths superior rolling stock travels Single, by with the mean value V of all path passage rates Average overallCompare, according to v Single〉=V Average overallCan obtain the path of " good " on the priori, it is included in " good " the path collection, and put in the path memory.
Above-mentioned Bayes classifier mainly is made of data divider, roadway characteristic maker, Bayes controller, road topology device and buffer;
A) " good " path is inputed to data divider, dispenser is divided into " good " path collection T1 of peak period and " good " path collection T2 of non-peak period according to the time attribute in path with it;
B) in the roadway characteristic maker,, the frequency of all paths through same road section e taken statistics to all " good " path collection;
C) in the roadway characteristic maker, calculate on the road network all nodes between shortest path, and obtain intermediary's centrality of road section e thus.
D) from road network data, select a part and come out, utilize the Bayes controller to come the road section classification is obtained preferential road collection, suboptimum road collection and Ordinary Rd collection as the training data of Bayes controller;
E) in the road topology detecting device,, the preferential road collection of this three class of its generation is put in the classification of road storer according to three road collection that step d) obtains.
The invention has the beneficial effects as follows: this patent is exactly to utilize this driver that the priori of the real-time traffic situation of road network is inferred, construct Bayes classifier and formed preferential identification to road network, a kind of path planning apparatus and the method that form based on this preferential identification, it can provide different optimal paths respectively according to the priori of peak period, non-peak period in different traffic slots.This device and method has good practicality, and can be integrated into flexibly in the various navigational system based on road traffic.
Description of drawings:
Accompanying drawing 1 is the principle schematic of device of the present invention;
Accompanying drawing 2 is synoptic diagram of path of the present invention screening washer;
Accompanying drawing 3 is the process flow diagrams in screening washer screening " good " path, path of the present invention;
Accompanying drawing 4 is synoptic diagram of Bayes classifier of the present invention;
Accompanying drawing 5 is Bayes schematic flow sheets of road collection of the present invention;
Accompanying drawing 6 is optimal path computation process flow diagrams of the present invention.
Embodiment:
The concrete embodiment of this device is as follows:
The enforcement of end of probe:
Obtain the trip empirical data with the GPS detector.Select n needs are the driver that the area road net of path analysis is familiar with very much, the GPS detector is contained on this n driver's the car, write down the trip data of its every day, wrap with GPS
Draw together the starting point and the terminating point in its trip path.Cumulative record is after their trip path a period of time (such as one month) like this, according to GPS detector and the record thereof on these driver's vehicles that Real-time Road is familiar with, obtains n experience driver's trip GPS track data.
Center processing end implementation step:
The data transmission of whole center processing end intermodule is controlled by central control unit, and the steps flow chart of operation is also controlled by central control unit.Provide Bayes's paths planning method that the center processing end is realized below sequentially.
1 usefulness path screening washer obtains the path collection of " good " on the priori, and this module mainly is made up of comparison controller, coupling computing module and buffer three parts, and its structural representation is seen " mistake! Do not find Reference source." its process flow diagram sees " mistake! Do not find Reference source.”
The GPS track data that central control unit obtains gathering from the GPS detector.
1a) in the coupling computing module, the driver's who collects according to end of probe the GPS track and the starting point and the terminating point of record, with map-matching method it is matched the road corresponding highway section, and join according to adjacent road section and to obtain the path that forms between each starting point and the terminating point.The path collection that the coupling computing module obtains is input in the comparison counter.
1b) in comparison controller, earlier according to the Hang journey Shi Jian ⊿ t between the total length L of every paths and this path are from the starting point to the terminating point, by v Single=L/ ⊿ t can obtain the average passage rate v of individual paths SingleCalculate all travel speeds of GPS in a certain period T then, according to the average V of the passage rate in all paths in this period Average overallSet a threshold value.Compare v SingleWith V Average overallRelation.If v Single〉=V Average overall, think that then this path is the path of " good " on the priori, includes it in " good " the path collection, and put in the path memory; If v Single﹤ V Average overall, think that then this path is not the path of " good " on the priori, gives up it.After forming the path collection of be over " good " at last, the notice central control unit deposits the path collection of these " good " in the path memory in.
2 usefulness Bayes classifiers are done classification to road network, obtain three kinds of preferred road collection.Bayes classifier mainly is made of these several parts of data divider, roadway characteristic maker, Bayes controller, road topology device and buffer, and its structure is as " mistake! Do not find Reference source " shown in, its flow process is as " mistake! Do not find Reference source " shown in.
" good " path collection that central control unit obtains from path memory is input to Bayes classifier.
2a) " good " path is inputed to data divider, dispenser is divided into " good " path collection T1 of peak period and " good " path collection T2 of non-peak period according to the time attribute in path with it.Following step 2b--2e implements T1 and T2 respectively, obtains the preferential road collection of peak period three classes different with non-peak period.
2b) in the roadway characteristic maker, to all " good " path collection, the frequency of all paths through same road section e taken statistics, promptly different paths is every through once identical road section, and its frequency attribute f (e) adds 1, and f (e) initial value is 0.Do on average through the speed of road section according to all " good " paths, be designated as v (e).
2c) in the roadway characteristic maker, calculate on the road network all nodes between shortest path, all shortest paths add up to N.According to the number of times c (e) of all shortest paths, calculate the centrality B of intermediary (e)=c (e)/N of highway section e through each road section e.
2d) selecting a part from road network data comes out as the training data of Bayes controller.According to 2b) and 2c) in feature f (e), the v (e) of the road e that obtains and the parameter that B (e) estimates the Bayes controller.Determined after the Bayes controller parameter, utilized the Bayes controller to come the road section classification is obtained preferential road collection E according to its feature f (e), v (e) and B (e) in all roads 1, suboptimum road collection E 2With Ordinary Rd collection E 3
2e) in the road topology detecting device, according to step 2d) three road collection obtaining, from E 1Beginning calculates E 1The maximum strong subnet C that is communicated with 1With E 1In not at C 1Road join E again 2In, calculate E 2The maximum strong subnet C that is communicated with 2Like this, we make G 1=C 1As preferential road collection, make G 2=C 1+ C 2As suboptimum road collection, G 3Be all road collection of whole road networks conducts.Like this, we just obtain by G 1, G 2, G 3The preferential road collection of three classes that forms.
Central control unit is put into the preferential road collection of this three class of its generation in the classification of road storer after the step of Bayes classifier is finished.
3 according to the preferential road collection of the classification at peak period T1 and non-peak period T2 that forms in the step 2, calculates optimal path respectively in the optimal path computation device.Because for the road collection of T1 and T2, its calculation procedure all is the same, like this, in the following description T1 and T2 is not distinguished.Its flow process is seen " mistake! Do not find Reference source.”
Central control unit is regarded the preferential road collection of three classes as the triple-decker network and is input to the optimal path computation device.Determine optimal path to be determined starting point and terminating point in road network.Several steps below in the optimal path computation device, implementing:
3a) from G iBegin forward and reverse search optimal path from the initial sum terminating point respectively in the layer;
If 3b) at G iIn layer forward and the reverse search, the node on the optimum subpath also is present in corresponding G I-1In the layer, be designated as S respectively I-1, T I-1, so just jump back to step 3a), and i subtracts 1, promptly at G I-1Middle execution in step 3a).If in forward and reverse search, there is not such node on the optimum subpath) so with regard to execution in step 3c;
If 3c) subpath of the subpath of forward lookup and reverse search can not meet at same point, continue execution in step 3b so), if may converge a bit, execution in step 3d so);
3d) subpath that converges every layer of calculating forms last experience optimal path.
In this device, end of probe equipment GPS detector is assemblied on selected driver's vehicle with the form of hardware, and can preserve the GPS track data of its vehicle.The center processing end device can be under any operating system platform, utilize any programming language, utilize software mode to realize, also can adopt suitable hardware to realize, have good practicality, and can be integrated into flexibly in the various navigational system based on road traffic.

Claims (4)

1.一种基于GPS探测的贝叶斯路径规划装置，其特征是：主要分为探测端和中心处理端两个部分，其中探测端选择一些熟悉道路交通状况的司机，将GPS探测器分布在这些司机的车辆上；所述的中心处理端包括接收器、中心控制器、路径筛选器、贝叶斯分类器和最优路径计算器几个模块： 1. A Bayesian path planning device based on GPS detection, characterized in that it is mainly divided into two parts, a detection end and a central processing end, wherein the detection end selects some drivers who are familiar with road traffic conditions, and distributes GPS detectors in On the vehicles of these drivers; the central processing end includes several modules of receiver, central controller, path filter, Bayesian classifier and optimal path calculator: 其中探测端： Among them, the detection terminal: GPS探测器，主要是用GPS采集器来对n个司机的行驶轨迹进行探测； The GPS detector mainly uses the GPS collector to detect the driving trajectories of n drivers; 所述的中心处理端： The central processing end: 中心控制器，控制其它的模块之间的数据传送； The central controller controls the data transmission between other modules; 路径筛选器，主要根据采集到的这些司机车辆的GPS轨迹点以及采集的出发点和终止点恢复出行驶的路径，并筛选出通过速度快、耗费时间短的“好”路径； The path filter mainly restores the driving path based on the collected GPS track points of these drivers' vehicles and the collected starting point and ending point, and filters out the "good" path with fast passing speed and short time consumption; 贝叶斯分类器，根据建立的“好”路径集训练贝叶斯分类器，根据训练好的贝叶斯分类器对道路做分类，形成优先道路集，次优道路集以及全部道路集； Bayesian classifier, train the Bayesian classifier according to the established "good" path set, and classify the roads according to the trained Bayesian classifier, forming a priority road set, a suboptimal road set and all road sets; 最优路径计算器，根据形成的三种优选道路，实施最优路径搜索计算，得到最优路径结果； Optimal path calculator, according to the formed three optimal roads, implement optimal path search and calculation, and obtain the optimal path result; 路径存储器，存放路径筛选器建立的“好”路径集； path memory, which stores the set of "good" paths established by the path filter; 道路分类存储器，存放贝叶斯分类器分类后形成的三种道路集。 The road classification storage stores the three road sets formed by the Bayesian classifier. 2.一种基于GPS探测的贝叶斯路径规划方法，其特征是包括以下步骤： 2. A Bayesian path planning method based on GPS detection, characterized in comprising the following steps: （1）、探测端用GPS探测器来获取出行经验数据； (1) The detection end uses GPS detectors to obtain travel experience data; （2）、中心控制端从GPS探测器中得到的轨迹数据经过一系列处理而实现路径规划，包括以下步骤： (2) The trajectory data obtained by the central control terminal from the GPS detector undergoes a series of processing to realize path planning, including the following steps: （2.1）用路径筛选器来得到先验上的“好”的路径集：主要根据采集到的这些司机车辆的GPS轨迹点以及采集的出发点和终止点恢复出行驶的路径，并筛选出通过速度快、耗费时间短的“好”路径； (2.1) Use a path filter to obtain a priori "good" path set: mainly based on the collected GPS track points of these drivers' vehicles and the collected starting and ending points to restore the driving path, and filter out the passing speed "good" paths that are fast and take less time; （2.2）根据建立的“好”路径集训练贝叶斯分类器，根据训练好的贝叶斯分类器对道路做分类，形成优先道路集，次优道路集以及全部道路集； (2.2) Train the Bayesian classifier according to the established "good" path set, classify the roads according to the trained Bayesian classifier, and form a priority road set, a suboptimal road set and all road sets; （2.3）最优路径计算器中计算最优路径，中心控制器将上述三类优先道路集看成三层道路网络输入到最优路径计算器，得到最优路径结果。 (2.3) The optimal path is calculated in the optimal path calculator, and the central controller regards the above three types of priority road sets as a three-layer road network and inputs them to the optimal path calculator to obtain the optimal path result. 3.根据权利要求2所述的基于GPS探测的贝叶斯路径规划方法，其特征是：所述的路径筛选器主要由比较控制器、匹配计算模块以及缓存器三部分组成： 3. the Bayesian path planning method based on GPS detection according to claim 2, is characterized in that: described path filter is mainly made up of comparison controller, matching calculation module and buffer three parts: a）在匹配计算模块中，根据探测端采集得到的司机的GPS轨迹和记录的出发点和终止点，用地图匹配方法将其匹配到对应的道路路段，并根据相邻的道路路段相接得到每个出发点和终止点之间形成的路径，匹配计算模块得到的路径集输入到比较计算器中； a) In the matching calculation module, according to the GPS track of the driver collected by the detection end and the recorded starting point and end point, use the map matching method to match it to the corresponding road section, and get each road section according to the adjacent road sections. The path formed between the starting point and the ending point, the path set obtained by matching the calculation module is input into the comparison calculator; b）在比较控制器中，先根据每条路径的总长度和GPS记录的行程时间，可以得到其平均通行速度v，通过比较平均通行速度的关系，得到一条先验上“好”的路径，将其纳入”好”的路径集，并放入到路径存储器中。 b) In the comparison controller, firstly, according to the total length of each path and the travel time recorded by GPS, its average passing speed v can be obtained, and by comparing the relationship between the average passing speed, a priori "good" path can be obtained, Include it in the "good" set of paths and put it in the path memory. 4.根据权利要求2所述的基于GPS探测的贝叶斯路径规划方法，其特征是：所述的贝叶斯分类器主要由数据分割器、道路特征生成器、贝叶斯分类控制器、道路拓扑器和缓存器构成； 4. the Bayesian path planning method based on GPS detection according to claim 2 is characterized in that: described Bayesian classifier is mainly composed of data segmenter, road feature generator, Bayesian classification controller, Road topology and buffer configuration; a）将“好”路径输入给数据分割器，分割器根据路径的时间属性，将其分为高峰期的“好”路径集T1和非高峰期的“好”路径集T2； a) Input the "good" path to the data splitter, and the splitter divides it into the "good" path set T1 during the peak period and the "good" path set T2 during the non-peak period according to the time attribute of the path; b）在道路特征生成器中，对所有的“好”路径集，将所有路径经过同一道路路段e的频次作统计； b) In the road feature generator, for all "good" path sets, make statistics on the frequency of all paths passing through the same road segment e; c）在道路特征生成器中，计算道路网络上的所有节点对之间的最短路径，并由此得到道路路段e的中介中心性； c) In the road feature generator, calculate the shortest path between all node pairs on the road network, and thus obtain the betweenness centrality of the road segment e; d）从道路网络数据中挑选一部分出来作为贝叶斯分类控制器的训练数据，利用贝叶斯分类控制器来对道路路段分类，得到优先道路集、次优道路集和一般道路集； d) Select a part of the road network data as the training data of the Bayesian classification controller, use the Bayesian classification controller to classify the road sections, and obtain the priority road set, the suboptimal road set and the general road set; e）在道路拓扑检测器中，根据步骤d）得到的三个道路集，将其生成的这三类优先道路集放入到道路分类存储器中。 e) In the road topology detector, according to the three road sets obtained in step d), put the generated three priority road sets into the road classification memory.
CN 201110106642 2011-04-27 2011-04-27 Bayes path planning device and method based on GPS (Global Positioning System) detection Expired - Fee Related CN102278995B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110106642 CN102278995B (en) 2011-04-27 2011-04-27 Bayes path planning device and method based on GPS (Global Positioning System) detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110106642 CN102278995B (en) 2011-04-27 2011-04-27 Bayes path planning device and method based on GPS (Global Positioning System) detection

Publications (2)

Publication Number Publication Date
CN102278995A true CN102278995A (en) 2011-12-14
CN102278995B CN102278995B (en) 2013-02-13

Family

ID=45104578

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110106642 Expired - Fee Related CN102278995B (en) 2011-04-27 2011-04-27 Bayes path planning device and method based on GPS (Global Positioning System) detection

Country Status (1)

Country Link
CN (1) CN102278995B (en)

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102809384A (en) * 2012-08-22 2012-12-05 北京世纪高通科技有限公司 Road link extraction method and device
CN103714708A (en) * 2013-12-18 2014-04-09 福建工程学院 Optimal path planning method based on split-time experience path of taxi
CN104062671A (en) * 2014-07-08 2014-09-24 中国石油大学(华东) Curvature-restrained GNSS floating car map matching method and device
CN104350473A (en) * 2012-06-06 2015-02-11 苹果公司 Context aware map application
CN104990559A (en) * 2015-07-27 2015-10-21 福建工程学院 Route recommending method based on taxi empirical data, system and client
CN107346478A (en) * 2016-05-04 2017-11-14 中国农业大学 Shipping paths planning method, server and system based on historical data
CN107766881A (en) * 2017-09-30 2018-03-06 中国地质大学（武汉） A kind of method for searching based on fundamental classifier, equipment and storage device
CN107941234A (en) * 2017-12-21 2018-04-20 广东欧珀移动通信有限公司 Travel route recommendation method and device, storage medium and mobile terminal
WO2018165848A1 (en) * 2017-03-14 2018-09-20 深圳市南北汽车美容有限公司 Method for analysing real-time road conditions to recommend travel time, and navigation system
CN110033642A (en) * 2019-04-18 2019-07-19 福建工程学院 A kind of indoor shared parking stall path experience generates and its bootstrap technique
CN110088573A (en) * 2016-12-20 2019-08-02 赫尔环球有限公司 For determining the equipment and associated method of vehicle location
CN111754800A (en) * 2020-04-13 2020-10-09 腾讯科技（深圳）有限公司 Method and device for acquiring road condition information and server
US10816349B2 (en) 2017-06-13 2020-10-27 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for route planning based on deep convolutional neural network
CN111829527A (en) * 2020-07-23 2020-10-27 中国石油大学(华东) A path planning method for unmanned ships based on deep reinforcement learning and considering marine environment elements
CN111858817A (en) * 2020-07-23 2020-10-30 中国石油大学(华东) A BiLSTM-CRF Path Inference Method for Sparse Trajectories
CN113887659A (en) * 2021-10-22 2022-01-04 沈阳世纪高通科技有限公司 Vehicle type classification algorithm based on naive Bayes

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1892182A (en) * 2005-07-01 2007-01-10 株式会社电装 Navigation system
US20070208492A1 (en) * 2006-03-03 2007-09-06 Inrix, Inc. Dynamic time series prediction of future traffic conditions
CN101126808A (en) * 2007-08-02 2008-02-20 中国科学院自动化研究所 A robot navigation system and navigation method
CN101916568A (en) * 2008-12-05 2010-12-15 索尼株式会社 Information processing apparatus, information processing method, and program

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1892182A (en) * 2005-07-01 2007-01-10 株式会社电装 Navigation system
US20070208492A1 (en) * 2006-03-03 2007-09-06 Inrix, Inc. Dynamic time series prediction of future traffic conditions
CN101126808A (en) * 2007-08-02 2008-02-20 中国科学院自动化研究所 A robot navigation system and navigation method
CN101916568A (en) * 2008-12-05 2010-12-15 索尼株式会社 Information processing apparatus, information processing method, and program

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李红连: "BP神经网络在车辆组合导航中的应用研究", 《压电与声光》 *

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104350473A (en) * 2012-06-06 2015-02-11 苹果公司 Context aware map application
CN104350473B (en) * 2012-06-06 2017-06-23 苹果公司 Context aware map application
US10274336B2 (en) 2012-06-06 2019-04-30 Apple Inc. Context aware map application
CN102809384A (en) * 2012-08-22 2012-12-05 北京世纪高通科技有限公司 Road link extraction method and device
CN102809384B (en) * 2012-08-22 2015-11-11 北京世纪高通科技有限公司 A kind of road chain extracting method and device
CN103714708A (en) * 2013-12-18 2014-04-09 福建工程学院 Optimal path planning method based on split-time experience path of taxi
CN104062671A (en) * 2014-07-08 2014-09-24 中国石油大学(华东) Curvature-restrained GNSS floating car map matching method and device
CN104990559A (en) * 2015-07-27 2015-10-21 福建工程学院 Route recommending method based on taxi empirical data, system and client
CN107346478A (en) * 2016-05-04 2017-11-14 中国农业大学 Shipping paths planning method, server and system based on historical data
CN110088573A (en) * 2016-12-20 2019-08-02 赫尔环球有限公司 For determining the equipment and associated method of vehicle location
WO2018165848A1 (en) * 2017-03-14 2018-09-20 深圳市南北汽车美容有限公司 Method for analysing real-time road conditions to recommend travel time, and navigation system
US10816349B2 (en) 2017-06-13 2020-10-27 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for route planning based on deep convolutional neural network
CN107766881A (en) * 2017-09-30 2018-03-06 中国地质大学（武汉） A kind of method for searching based on fundamental classifier, equipment and storage device
CN107766881B (en) * 2017-09-30 2020-06-26 中国地质大学（武汉） Pathfinding method, device and storage device based on basic classifier
CN107941234A (en) * 2017-12-21 2018-04-20 广东欧珀移动通信有限公司 Travel route recommendation method and device, storage medium and mobile terminal
CN110033642A (en) * 2019-04-18 2019-07-19 福建工程学院 A kind of indoor shared parking stall path experience generates and its bootstrap technique
CN111754800A (en) * 2020-04-13 2020-10-09 腾讯科技（深圳）有限公司 Method and device for acquiring road condition information and server
CN111829527A (en) * 2020-07-23 2020-10-27 中国石油大学(华东) A path planning method for unmanned ships based on deep reinforcement learning and considering marine environment elements
CN111858817A (en) * 2020-07-23 2020-10-30 中国石油大学(华东) A BiLSTM-CRF Path Inference Method for Sparse Trajectories
CN111858817B (en) * 2020-07-23 2021-05-18 中国石油大学(华东) A BiLSTM-CRF Path Inference Method for Sparse Trajectories
CN111829527B (en) * 2020-07-23 2021-07-20 中国石油大学(华东) A path planning method for unmanned ships based on deep reinforcement learning and considering marine environment elements
CN113887659A (en) * 2021-10-22 2022-01-04 沈阳世纪高通科技有限公司 Vehicle type classification algorithm based on naive Bayes
CN113887659B (en) * 2021-10-22 2024-07-12 沈阳世纪高通科技有限公司 Vehicle type classification algorithm based on naive Bayes

Also Published As

Publication number Publication date
CN102278995B (en) 2013-02-13

Similar Documents

Publication Publication Date Title
CN102278995A (en) Bayes path planning device and method based on GPS (Global Positioning System) detection
WO2022166239A1 (en) Vehicle travel scheme planning method and apparatus, and storage medium
CN106525058B (en) A kind of vehicle group trip air navigation aid and device
KR20210032283A (en) Extended object tracking using radar
CN106251642B (en) A kind of public transport road chain speed calculation method based on real-time bus GPS data
CN107958302A (en) Empirical path planing method based on virtual topology transportation network
CN106441324A (en) Navigation method, navigation apparatus, road condition information collection method and road condition information collection system
CN108510737B (en) Unmanned vehicle power supply real-time monitoring method and device integrating wind environment
CN110032381B (en) Method, system and storage medium for updating road state of electronic map
CN108335483A (en) The estimating method and its system of traffic congestion diffusion path
CN114038214B (en) Urban traffic signal control system
CN106157619B (en) Road network is in fortune vehicle number calculating method under hypersaturated state
Elbery et al. Eco-routing: an ant colony based approach
CN109410584A (en) A kind of road conditions detection method and device
Geng et al. Human-driver speed profile modeling for autonomous vehicle's velocity strategy on curvy paths
CN111710160A (en) A Travel Time Prediction Method Based on Floating Car Data
CN105806355B (en) A kind of vehicle green path navigation system and method
CN117537836A (en) A navigation path optimization method considering real-time accident risk
JP2013206325A (en) Traffic information prediction device, traffic information prediction method and computer program
CN110782656A (en) A road bottleneck point identification method, device, electronic device and storage medium
Yu et al. A review of the link traffic time estimation of urban traffic
CN104697543B (en) A kind of path searching method therefor for merging individual character preference heterogeneity
JP2011076410A (en) Device, computer program, and method for estimating traffic information
Habtie et al. Cellular network based real-time urban road traffic state estimation framework using neural network model estimation
Habtie et al. Road traffic state estimation framework based on hybrid assisted global positioning system and uplink time difference of arrival data collection methods

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20130213

Termination date: 20180427

CF01 Termination of patent right due to non-payment of annual fee