WO2017158391A1 - Processus de planification d'itinéraire le plus rapide pour véhicules routiers - Google Patents

Processus de planification d'itinéraire le plus rapide pour véhicules routiers Download PDF

Info

Publication number
WO2017158391A1
WO2017158391A1 PCT/HR2016/000011 HR2016000011W WO2017158391A1 WO 2017158391 A1 WO2017158391 A1 WO 2017158391A1 HR 2016000011 W HR2016000011 W HR 2016000011W WO 2017158391 A1 WO2017158391 A1 WO 2017158391A1
Authority
WO
WIPO (PCT)
Prior art keywords
portal
trajectory
graphs
destination
graph
Prior art date
Application number
PCT/HR2016/000011
Other languages
English (en)
Inventor
Tonci TOMIC
Original Assignee
Mireo D.D.
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 Mireo D.D. filed Critical Mireo D.D.
Priority to PCT/HR2016/000011 priority Critical patent/WO2017158391A1/fr
Publication of WO2017158391A1 publication Critical patent/WO2017158391A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes

Definitions

  • This invention resolves the computer process of planning the fastest route for road vehicles from a starting to a destination point.
  • the subject of the invention is classified under G06N Computer systems based on specific computational models, or more precisely in class G06N 5/00 Computer systems utilizing knowledge based models, and can alternatively be classified in class G01 C 21/26 Navigational instruments specially adapted for navigation in a road network.
  • the problem addressed with this invention is maintaining the accuracy of a route calculation when planning the fastest route for road vehicles from a starting to a destination point in the event of a large amount of real data.
  • the method according to this invention uses the combination (A) of a graph that contains a subset of all major roads with intersections as nodes, where the intersections in the nodes do not have prohibited turns, and two ad hoc graphs (B 1 ) and (B2), which are created in the area surrounding the selected start point (start) and end point (destination).
  • Dijkstra's algorithm is based on the principle that all possible paths are traveled, starting from the source node to neighboring nodes. When arriving at a neighboring node, it is assigned a magnitude from the source node increased by the magnitude of the path. It is obvious that in this manner many various paths can be used to arrive at any node along the path.
  • the previously calculated magnitude of that node is compared with the newly calculated magnitude. If the new magnitude is greater than or equal to the previous magnitude, the new path is ignored; if it is less, the previous path is rejected and replaced with the new path.
  • the recent improvement to Dijkstra's algorithm the contraction hierarchies (CH) algorithm, operates in the manner that individual routes are calculated in advance and are saved as shortcuts. Therewith, the number of nodes is decreased, which accelerates the operation of Dijkstra's algorithm.
  • the objective of this invention is to retain the accuracy of the calculation of a route when planning the fastest route for road vehicles from a starting to a destination point, while maintaining the performance of the calculation.
  • Dijkstra's algorithm operates on a directed graph. Intersections can be designated as nodes in the graph, and the roads can be designated as paths - as shown in figure la and figure lb. Each path in the graph has its magnitude. Magnitude, in this case, is proportional to the driving time from the beginning to the end of the road segment.
  • An alternative method of constructing the graph according to figure 1 c treats road segments as nodes, and paths become intersections which can be accessed from segment to segment. Such a graph is called a dual graph.
  • a dual graph created from a road network is always more complex than a graph that designates intersections as nodes and therefore, its use is avoided when working with a large amount of data.
  • Dijkstra's algorithm sometimes may not provide an optimal result, because it is not possible to incorporate information into the algorithm regarding which path was used to arrive at a particular node (figure Id).
  • the problem of prohibited turns does not exist; for the given example, the path eEg will simply be eliminated.
  • Figures la, lb, lc and Id show the calculation method of the route noted in the state of the art, whereby Figure la shows two intersections connected with a two-way road; Figure lb shows a graph of intersections ABCDEF interconnected with the rectangular road network abcdefg, whereby all roads are two-way except for road "g", which connects intersections "E” and "B”; Figure 1 c shows the dual graph of the initial graph; and Figure Id shows the dual graph with information regarding the prohibited turn at intersection E.
  • Figure 2 displays the schematic representation of the hybrid calculation of the route using a combination of edge-based routing on ad hoc graphs B 1 and B2 and node- based routing on the CH graph.
  • Figure 3 shows the method of calculating the shortest route in the flow diagram.
  • Figure 4 provides an example of a schematic representation of the calculation method on an actual map with portals with the method of calculating the route.
  • Figure 5a provides an example of calculating a route from A to B with a U-turn on an actual map.
  • Figure 5b shows a locally resolved U-turn problem at portal X.
  • Figure 6a shows the problem of closing routes in the event of graphs with a small number of portals per intersection.
  • Figure 2 displays a CH graph (contraction hierarchies graph) which contains a subset of all major roads.
  • Points al, a2, a3, bl, b2, b3 and b4 are nodes of the CH graph and function as portals from which the switch is made to the node-based routing graph. In this case, portals do not contain prohibited turns.
  • ad hoc graphs Bl and B2 are created.
  • the stated graphs B l and B2 have roads as nodes; route calculation using Dijkstra's algorithm with these graphs functions perfectly.
  • Graphs Bl and B2 are connected through the portals with the CH graph.
  • Each portal has its corresponding magnitude, which is obtained by calculating the time necessary to arrive to the start of portal a, or from portal b to the destination. Then, the fastest path from one of the portals a to one of the portals b is calculated based on the CH graph, taking into account the initial magnitude of the portal.
  • the final shortest route represents the chain start - a - b - destination that has the lowest magnitude, i.e. when the sum of the corresponding trajectories from the graphs Bl + B2 + CH is the smallest.
  • Figure 3 shows the process of calculating the shortest route in a flow diagram, where it can be noted that the construction of the CH graph is conducted first in the calculation process. Then the ad hoc graph is created around the starting point, and entry portals are gathered on the ad hoc graph. An ad hoc is then also created around the point and exit portals are gathered. Then the routing process is executed on the CH graph and finally, the correction of U-turns is executed.
  • FIG 4 the creation of the ad hoc graph that originates in the center of the figure is noted.
  • the points on the figure represent portals (connection points with the CH graph).
  • Marked roads represent a part of the graph.
  • the points on marked roads are portals that will be associated with the corresponding CH graph.
  • the process begins with the creation of the ad hoc graph from the position marked with a cross (x) in its center. Expansion is executed in all possible directions from the starting point, including through each intersection that is encountered. Expansion continues until all trajectories are reached or until a point designating a portal is reached or until a dead end street is reached. After encountering a portal, expansion is executed precisely another two steps in order to most efficiently close open trajectories that could meander around the portal.
  • FIG. 5a The problem with the existence of a U-turn is shown in figure 5a, which for the route from A to B, contains a U-turn at portal X.
  • the location at portal X is encountered.
  • the same portal X is encountered by expansion in the direction opposite of point B.
  • the CH route will aggregate in point X, so that the complete route will be A - X - B.
  • arrival at point X and departure from that same point makes use of the same road but in opposite directions, resulting in a U-turn being present at that location. Deeper analysis allows the conclusion that the cause of the U-turn is actually a prohibited turn shown in the figure.
  • the U-turn problem is resolved as shown in figure 5b. Namely, when encountering a U-turn during the route calculation, such is resolved by executing another ad hoc graph with roads acting as nodes in the vicinity of point X. A local route is calculated based on that graph, where the starting point is the road in the direction of point X, and the destination point is that same road, but this time in the opposite direction. The stated local route is then "glued" to the connection X.
  • Figure 6 shows the strategy "stop at the portal", which represents a problem in graphs with a small number of portals per intersection, which occurs in the case of the accumulation of a large number of minor roads.
  • Figure 6a shows an example of what can occur if expansion is stopped when encountering a portal. It is important to notice the scale in the upper left corner, and that it was necessary to create an enormous local graph in order to terminate all trajectories with portals. This situation occurs when a dense network of minor roads exists, which is why portals are very sparsely distributed.
  • the graph is prepared using the contraction hierarchies (CH) method in the vicinity of the starting point (start) and the end point (destination) and based on the basic data of the display map, ad hoc graphs Bl and B2 ( Figure 2) are created with roads designated as nodes in order to eliminate the problem of prohibited turns, and the further calculation of the fastest route is executed on the ad hoc graph by performing an expansion from the starting point in all possible directions and the expansion is continued until an entry portal (Pu) is reached on each trajectory, and expansion is also executed from the end (destination) point in all possible directions from the stated destination point and the expansion is continued until an exit portal (Pi) is reached on each trajectory
  • the calculation is further continued until all possible trajectories are closed.
  • Each trajectory is assigned a corresponding magnitude.
  • the magnitude of a trajectory is the sum of the road magnitudes that comprise that trajectory.
  • the magnitude of a portal can also be defined.
  • One or more trajectories pass through a portal.
  • the magnitude of a portal is the magnitude of the "lightest" trajectory in that point Bl m i n .
  • the described process of planning the fastest route for road vehicles can be used with all electronic devices that allow navigation of road vehicles, or those devices that use any type of navigation system between two points, such as tablets, mobile phones, GPS devices, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Navigation (AREA)

Abstract

Le problème qui est résolu avec cette invention est de maintenir la précision de calcul d'itinéraire lors de la planification de l'itinéraire le plus rapide pour des véhicules routiers entre un point de départ et un point de destination, tout en conservant la performance de calcul. Le processus selon la présente invention utilise une combinaison de graphiques qui contiennent un sous-ensemble de toutes les routes principales avec des intersections désignées comme des nœuds, mais sans intersections qui contiennent des virages interdits, et deux graphiques ad hoc qui sont créés au voisinage du point de départ (départ) et du point final (destination) sélectionnés. Le processus de planification de l'itinéraire le plus rapide pour des véhicules routiers comprend la création, au voisinage du point de départ et du point de destination, de deux graphiques ad hoc B1 et B2. Les graphiques B1 et B2 mentionnés mettent en œuvre des routes en tant que nœuds et exécutent des calculs à l'aide d'un algorithme de Dijkstra, qui fonctionne parfaitement avec les graphiques sujets. Les graphiques B1 et B2 sont reliés à des portails avec un graphique CH. Chaque portail a sa propre amplitude correspondante, qui est obtenue par calcul du temps nécessaire pour arriver du départ au portail a, ou du portail b à la destination. Sur la base du graphique CH, le trajet le plus rapide est ensuite calculé à partir de l'un des portails a à l'un des portails b, tenant compte de l'amplitude initiale des portails. L'itinéraire final le plus court représente la chaîne départ-a-b-destination, qui a la plus faible amplitude, c'est-à-dire lorsque la somme des trajectoires correspondantes à partir des graphiques B1 + CH + B2 est la plus faible.
PCT/HR2016/000011 2016-03-17 2016-03-17 Processus de planification d'itinéraire le plus rapide pour véhicules routiers WO2017158391A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/HR2016/000011 WO2017158391A1 (fr) 2016-03-17 2016-03-17 Processus de planification d'itinéraire le plus rapide pour véhicules routiers

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/HR2016/000011 WO2017158391A1 (fr) 2016-03-17 2016-03-17 Processus de planification d'itinéraire le plus rapide pour véhicules routiers

Publications (1)

Publication Number Publication Date
WO2017158391A1 true WO2017158391A1 (fr) 2017-09-21

Family

ID=55802399

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/HR2016/000011 WO2017158391A1 (fr) 2016-03-17 2016-03-17 Processus de planification d'itinéraire le plus rapide pour véhicules routiers

Country Status (1)

Country Link
WO (1) WO2017158391A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107860386A (zh) * 2017-10-17 2018-03-30 洛阳中科龙网创新科技有限公司 一种基于Dijkstra算法的农用机械最短路径规划的方法
CN107920379A (zh) * 2017-10-21 2018-04-17 天津大学 一种基于增加传输距离和重传的能量优先路由方法
CN110162033A (zh) * 2018-03-25 2019-08-23 环达电脑(上海)有限公司 路线规划和处理禁止的复杂驾驶操控的方法
CN113033907A (zh) * 2021-04-07 2021-06-25 上海钛米机器人股份有限公司 路径规划方法及装置、电子设备、存储介质

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050010358A1 (en) * 2001-12-17 2005-01-13 Societe De Technologie Michelin Route determination method and device
EP2645063A1 (fr) * 2012-03-28 2013-10-02 Fujitsu Limited Procédé de recherche de chemin et dispositif de recherche de chemin

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050010358A1 (en) * 2001-12-17 2005-01-13 Societe De Technologie Michelin Route determination method and device
EP2645063A1 (fr) * 2012-03-28 2013-10-02 Fujitsu Limited Procédé de recherche de chemin et dispositif de recherche de chemin

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LINGKUN WU ET AL: "Shortest Path and Distance Queries on Road Networks: An Experimental Evaluation", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 31 January 2012 (2012-01-31), XP080561361 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107860386A (zh) * 2017-10-17 2018-03-30 洛阳中科龙网创新科技有限公司 一种基于Dijkstra算法的农用机械最短路径规划的方法
CN107860386B (zh) * 2017-10-17 2020-09-04 洛阳中科龙网创新科技有限公司 一种基于Dijkstra算法的农用机械最短路径规划的方法
CN107920379A (zh) * 2017-10-21 2018-04-17 天津大学 一种基于增加传输距离和重传的能量优先路由方法
CN110162033A (zh) * 2018-03-25 2019-08-23 环达电脑(上海)有限公司 路线规划和处理禁止的复杂驾驶操控的方法
CN113033907A (zh) * 2021-04-07 2021-06-25 上海钛米机器人股份有限公司 路径规划方法及装置、电子设备、存储介质

Similar Documents

Publication Publication Date Title
US9322660B2 (en) Incremental map generation, refinement and extension with GPS traces
CN111133277A (zh) 使用时间和空间生成路线的方法、设备和计算机程序产品
WO2017158391A1 (fr) Processus de planification d'itinéraire le plus rapide pour véhicules routiers
Yu et al. A multi-modal route planning approach with an improved genetic algorithm
EP3561453A1 (fr) Procédé, appareil et produit programme informatique permettant de déterminer la probabilité d'un itinéraire
CN109491378A (zh) 自动驾驶车辆的基于道路分段的路线引导系统
Apple et al. Green driver: Ai in a microcosm
US10168176B2 (en) Visualizing unidirectional traffic information
US10928211B2 (en) Method and apparatus for selectively qualifying trajectories in regards to a determination of travel time for a maneuver
Andreev et al. Towards realistic pedestrian route planning
CN108168572A (zh) 用于地图导航的地图显示方法及装置
CN104697543B (zh) 一种融合个性偏好因子的路径探索方法
JP4736590B2 (ja) 候補経路作成装置、方法、プログラム、交通シミュレーション装置、方法及びプログラム、経路探索装置、方法、及びプログラム
JP7135699B2 (ja) 経路探索システムおよび経路探索プログラム
Fu et al. Route planning for OSU-ACT autonomous vehicle in DARPA urban challenge
CN105427581B (zh) 基于浮动车数据的交通仿真方法及系统
JP7173310B2 (ja) 経路探索装置、経路探索方法、及び経路探索プログラム
CN114783200A (zh) 一种动静态联动的车辆路径引导方法、装置及存储介质
Jung et al. Lane level path planning for urban autonomous driving using vector map
JP2009156940A (ja) 経路探索方法
JP4127774B2 (ja) 歩行者ナビゲーションプログラム
JPH09166453A (ja) 経路誘導装置
Li et al. Dynamic Programming Approaches for Solving Shortest Path Problem in Transportation: Comparison and Application
Arvidsson et al. Real-time route prediction of emergency vehicles
Pazil et al. Shortest Path from Bandar Tun Razak to Berjaya Times Square using Dijkstra Algorithm

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16717693

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 16717693

Country of ref document: EP

Kind code of ref document: A1