CN103344248A - Optimal path calculation method for vehicle navigation system - Google Patents

Optimal path calculation method for vehicle navigation system Download PDF

Info

Publication number
CN103344248A
CN103344248A CN2013102976235A CN201310297623A CN103344248A CN 103344248 A CN103344248 A CN 103344248A CN 2013102976235 A CN2013102976235 A CN 2013102976235A CN 201310297623 A CN201310297623 A CN 201310297623A CN 103344248 A CN103344248 A CN 103344248A
Authority
CN
China
Prior art keywords
node
path
nodes
processlist
step
Prior art date
Application number
CN2013102976235A
Other languages
Chinese (zh)
Other versions
CN103344248B (en
Inventor
朴燕
王晗
Original Assignee
长春理工大学
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 长春理工大学 filed Critical 长春理工大学
Priority to CN201310297623.5A priority Critical patent/CN103344248B/en
Publication of CN103344248A publication Critical patent/CN103344248A/en
Application granted granted Critical
Publication of CN103344248B publication Critical patent/CN103344248B/en

Links

Abstract

The invention relates to an optimal path calculation method for a vehicle navigation system. The method refers to an optimal path algorithm for the vehicle navigation system, the turning point of a grid map is identified and selected only in the execution process, the intermediate node from one turning point to an adjacent turning point can be ignored, and the number of nodes required to be treated is greatly reduced. The provided algorithm is simple, the grid map is not required to be subjected to pretreatment and extra computer storage space overhead, the optimal path can be efficiently searched, the vehicle navigation efficiency is improved, the vehicle path can be well planned in a complex map environment, and the optimal path is rapidly searched for the vehicle.

Description

一种车辆导航系统的最佳路径计算方法技术领域 Optimal route in a vehicle navigation system calculating FIELD

[0001] 本发明涉及一种车辆导航系统的最佳路径算法,特别涉及一种基于TPA*算法的复杂地图环境下车辆导航系统的最佳路径寻找方法。 [0001] The present invention relates to a vehicle navigation system optimal path algorithm, particularly to a method to find the optimal path of the vehicle navigation system in the complex environment map TPA * algorithm.

背景技术 Background technique

[0002] 车辆导航系统中最主要的部分是确定最佳路径的算法,用来确定车辆在道路交通网中从当前位置到目的地的最佳路径。 [0002] The main part of the vehicle navigation system is to determine the best path algorithm is used to determine the vehicle from the current position to the best path to the destination in road traffic network. 对于大多数车辆导航系统来说,起始点和终点之间的最佳路径通常定义为预计行进时间最短的路径。 For most vehicle navigation systems, the best path between the starting point and end point is usually defined as the estimated travel time to the shortest path. 确定车辆导航系统中最佳路径的主要思路是:用最短路径算法计算出车辆当前位置到目的地的完整路径,使得其预计旅行时间最短,最后把计算出的路径提供给车辆驾驶员参考。 The main idea of ​​determining a vehicle navigation system is the best path: to calculate the vehicle's current position with the shortest path algorithm full path to the destination, making it the shortest estimated travel time, and finally calculated path to the driver of the vehicle to provide a reference.

[0003] 最佳路径是寻找网络中节点对直接最短路径的一类问题,通常是采用图论和数学规划的方法进行对路径的搜索。 [0003] is the best path to find a class of problems nodes direct shortest path network, usually using graph theory and mathematical programming method to search for the path. 目前,最佳路径问题已经在计算机科学、运筹学、地理信息科学等众多领域广泛应用。 Currently, the best path problem has been widely used in many fields of computer science, operations research, geographic information science. 在具体应用上包括智能交通系统、地理信息系统、路径规划问题、计算机网络通信等。 Including intelligent transportation systems, geographic information systems, path planning, computer network communications on the specific application.

[0004]自上世纪50年代开始,各领域的学者对最佳路径问题进行了深入的研究。 [0004] Since the 1950s, scholars in various fields on the best path issues in-depth study was carried out. 基于Bellman优化原理,求解节点网络中一个节点到所有其他节点最佳路径的Dijkstra算法,求解节点网络里任意对节点之间最短路径的Floyd算法,以及智能搜索A*算法等为代表的经典算法奠定了节点网络最佳路径算法的基础。 Based on Bellman optimization principle, Dijkstra's algorithm for solving a node in the network node the best path to all other nodes in the network node to solve any of Floyd algorithm for the shortest path between nodes, and intelligent search algorithm A * algorithm as the representative of the classic laid the basis of the best path algorithm node network.

[0005] 在众多的最佳路径算法中,A*算法是目前比较流行的一种最佳路径的搜索算法,同Dijkstra等算法相比,A*算法采用了启发式的搜索策略,在搜索的过程中对每一个搜索的节点进行估计和计算,避免了路径搜索的盲目性,可以很大程度上减少搜索过程中遍历的节点,从而提高算法的执行速度。 [0005] Among the best path algorithm, A * algorithm is more popular search algorithm for optimal path, compared with algorithms such as Dijkstra, A * algorithm uses a heuristic search strategy, in search of during the course of the search for each node estimates and calculations, to avoid the blindness of the search path, it can largely reduce the node traversal of the search process, thereby increasing the execution speed of the algorithm.

[0006] 通过对目前最佳路径算法性能的比较可以发现,当应用在大型复杂的节点网络中的时候,即使性能比较优越的A*算法也存在着遍历节点数量多、运算速度慢等问题,在应用于复杂地图环境下的车辆导航系统中时,也不能很好地满足车辆导航系统对实时性的要求,因此实有必要提出改进的算法来解决复杂地图环境下车辆导航的问题。 [0006] Through the current best path comparison algorithm performance can be found when applied in large, complex nodes in the network, even though more superior properties of A * algorithm has a number of node traversal, the operation speed to be slow. when used in vehicle navigation system in the complex map of the environment, but also can not meet the vehicle navigation system for real-time requirements, and therefore the real need to propose an improved algorithm to solve vehicle navigation map in complex environmental problems.

发明内容 SUMMARY

[0007] 本发明提出一种车辆导航系统的最佳路径计算方法,以解决复杂地图环境下车辆导航的问题,通过识别和减少以最佳的路径到达终点所需要遍历的节点来提高算法的速度,具有算法实现简单,不占用计算机额外的存储空间,不需要对地图的预处理等优点,能快速地在复杂地图环境下寻找到最佳路径,对车辆进行导航。 [0007] The present invention proposes a vehicle navigation system, the optimal path calculation method to solve complicated map in the vehicle navigation environment, the best path to reach the end point by identifying and reducing the need to traverse the nodes to increase the speed of the algorithm with algorithm is simple and does not take up additional computer storage space, etc. without pretreatment of the map, to quickly find the best path in a complex environment map, the vehicle navigation.

[0008] 本发明采取的技术方案是,包括以下步骤: [0008] The present invention adopts the technical solution, comprising the steps of:

[0009] 步骤一:建立二维无方向、权重相同的网络节点栅格地图,其中,每个网络节点n的邻域节点集合neighbors (n)中都包含有8个节点,根据车辆行驶地图不同地理特征,每个节点有可通行和不可通行两种状态,一个节点通过直线移动和斜线移动两种方式到达另外一个相邻的节点,相邻节点间的距离默认为I; [0009] Step a: establishing two-dimensional non-directional, the same weight map grid network nodes, wherein each network node n neighbors neighborhood set of nodes (n) are included in the eight nodes, in accordance with the vehicle running maps for geographic features, each node can access and impassable two states, reaches a node adjacent to a node further linear movement and oblique movement in two ways, the distance between adjacent nodes by default I;

[0010] 步骤二:对节点n的邻接节点集合neighbors (n)进行删减,目的是要确定出同到达目标的最佳路径不相关的节点x,根据父节点P(n)到节点n的方向不同而需要考虑两种情况,分别是直线移动和斜线移动;另外,如果节点n是起点,因为其父节点为空,故而无法进行邻接节点的删减; [0010] Step two: the set of neighbors of the adjacent node (n) node n for exclusion, the purpose is to determine the best path to reach the same destination node uncorrelated x, according to the parent node P (n) to the node n different directions need to consider two cases, namely, linear movement and oblique movement; Further, if the node n is the starting point, because its parent node is empty, and therefore can not be cut adjacent node;

[0011] (I)、直线移动 [0011] (I), the linear movement

[0012] 在这种情况下,对满足公式(I)中约束条件的任何节点X (XG neighbors (n))进行删减: [0012] In this case, satisfying the formula (I) for the deletion of any constraint node X (XG neighbors (n)):

[0013] Ien (〈p (n),…,x>\n) < Ien (〈p (n),n, x>) (I) [0013] Ien (<p (n), ..., x> \ n) <Ien (<p (n), n, x>) (I)

[0014] (2)、斜线移动 [0014] (2), oblique movement

[0015] 同上述直线删减规则相似,但是在斜线移动的情况下,对节点X的删减限制条件更加的严格,需要满足公式(2)中的约束条件: [0015] with the deletion rule is similar to the straight, but in the case of oblique movement, deletion restrictions on node X more strictly necessary to satisfy equation (2) the constraints:

[0016] Ien «p (n), •••, x>\n) <len «p (n), n, x» (2) [0016] Ien «p (n), •••, x> \ n) <len« p (n), n, x »(2)

[0017] 在上述的两种情况中,neighbors (n)中都没有包含障碍节点,将此时经过删减之后得到的节点称为节点n的普通邻接节点(Normal Neighbor);而当neighbors (n)中含有障碍节点时,会出现不能完全对非普通邻接节点进行删减的情况,这时称这些未被删减掉的节点为特殊节点(Special Neighbor);如果一个节点x (x G neighbors (n))是特殊节点,那么它应该满足:节点X不是节点n的普通节点,并且满足公式(3)中的约束条件。 [0017] In the above two cases, neighbors (n) are not included obstacle node, the deletion obtained after this time is called common node adjacent node (Normal Neighbor) node n; and when neighbors (n when) contains obstacles node, there will not be entirely the case of non-ordinary adjacent node for exclusion, then call these nodes are not cut out for a particular node (special Neighbor); if a node x (x G neighbors ( n)) is a particular node, then it should satisfy: X is not a common node of the node n, and satisfies the formula (constraint 3).

[0018] Ien «p (n), n, x» <len «p (n), •••, x>\n) (3) [0019] 步骤三:进行对路径转折点的搜索工作,节点n到节点XG neighbor (n)的移动方向泾是一个关键因素,对转折点的定义如下: [0018] Ien «p (n), n, x» <len «p (n), •••, x> \ n) (3) [0019] Step Three: a turning point for the path searching operation, node n to node XG neighbor (n) in the direction of movement is a key factor Jing, turning points defined as follows:

[0020] 节点m是节点n的后继节点,n到m的方向为],如果n在此方向上经过最少k次单位移动后到达m,n\im = n + kd,并且满足下列约束条件之一,则称节点m是节点n的转折 [0020] The node m is the successor node n,, n to m in the direction], if n least k times the unit is moved in this direction after reaching the m, n \ im = n + kd, and satisfies the following constraints of First, the node is said node n m is turning

占.[0021] I).节点m是终点; Accounting for [0021] I) m is the end node.;

[0022] 2).节点m的邻接节点中至少有一个是特殊节点; [0022] 2) adjacent to the node m is at least a particular node;

[0023] 3).3是斜线方向时,存在一个节点/>=/w +矣< 在^ GpZ1JZ2)方向上距离m节点 [0023] 3) .3 is oblique direction, the presence of a node /> = / w + carry <m from nodes on ^ GpZ1JZ2) direction

h个单位距离,并且节点p是满足上述条件I或条件2的一个转折点; h units of distance, and the node p is the condition satisfying the above conditions, or a turning point I 2;

[0024] 步骤四:将按照步骤三中定义得到的、属于节点n的转折点存放到数组ProcessList中,任意节点x G ProcessList是待处理的节点,在每一次向ProcessList中插入节点的时候,都需要保持ProcessList的有序性,即ProcessList中的第一个节点的路径开销值:即从起点到此节点的行进距离最小,最后一个节点的路径开销值最大,在下一次算法循环的时候,取出ProcessList中的第一个节点进行处理; [0024] Step Four: defined according to step 3 to obtain the inflection point belonging to the node n stored in array the ProcessList, any node x G ProcessList the node to be processed, when each of the insertion node to the ProcessList, need ProcessList orderly retention, i.e. the path the first node in the ProcessList cost values: i.e. the minimum travel distance from the start point to this node, the last node of the maximum value of the path cost, the cycle time of the next algorithm, remove the ProcessList a first processing node;

[0025] 步骤五:循环执行上述步骤二至步骤四,当终点ngMl被插入到ProcessList中时,完成对最佳路径的搜索,结束;从终点ngMl开始,依次将其前继节点即父节点添加到路径数组PathList中,直到将起点nstart添加到PathList中,路径数组PathList中保存的即为寻找到的最佳路径上的所有路径节点,车辆导航系统可以按照此最佳路径对行驶在地图上的车辆进行实时的导航工作。 [0025] Step Five: loop above steps two to step four, when the end is inserted into the ProcessList ngml complete the search for the best path, ending; ngml starting from the end, i.e., the node will continue sequentially added before the parent node the array of paths PathList until the start of nstart added to PathList, all nodes on the path to save the best path is the path to find an array PathList in the vehicle navigation system can follow this best path to travel on the map real-time vehicle navigation work. [0026] 与现有的算法相比,本发明一种车辆导航系统的最佳路径算法在执行的过程中只对栅格地图中的转折点进行识别和选取,从一个转折点到其相邻的下一个转折点的中间节点可以被忽略,进而大量地减少了需要处理的节点数量。 [0026] Compared with the conventional algorithms, the optimal path algorithm in a vehicle navigation system according to the present invention, the turning point of the grid map to identify and select only the implementation of the process, from which a turning point to the next adjacent a turning point in the intermediate node may be ignored, thereby substantially reduced the number of nodes to be processed. 本发明提出的算法简单,不需要对栅格地图进行任何的预处理和额外的计算机存储空间开销,可以高效率的搜索出最佳路径,提高车辆导航的效率。 The proposed algorithm of the present invention is simple and does not require any pretreatment grid map and additional computer memory space overhead, can efficiently search the best route, the vehicle navigation efficiency. 采用本发明,可以很好的在复杂地图环境下进行对车辆路径的规划,迅速为车辆寻找最佳路径。 According to the present invention can be a good vehicle path planning in a complex environment map, quickly find the best path for the vehicle.

附图说明 BRIEF DESCRIPTION

[0027] 图1是本发明算法中涉及到的等长路径的示意图; [0027] FIG. 1 is a schematic diagram of the long path algorithms present invention relates to;

[0028] 图2是本发明算法中在不同情况下对邻域节点进行删减的示意图; [0028] FIG. 2 is a schematic diagram of the neighbor nodes for exclusion algorithm of the present invention in different situations;

[0029] 图3是本发明算法中对转折点的搜索示意图。 [0029] FIG. 3 is a schematic diagram of the search algorithm of the present invention for turning points.

具体实施方式: Detailed ways:

[0030] 以下通过特定的具体实例并结合附图说明本发明的实施方式,本领域技术人员可由本说明书所揭示的内容轻易地了解本发明的其它优点与功效。 [0030] The following description of embodiments and drawings of the present invention, by binding certain specific examples, those skilled in the art may be disclosed in the present specification easily understand other advantages and effects of the present invention. 本发明亦可通过其它不同的具体实例加以施行或应用,本说明书中的各项细节亦可基于不同观点与应用,在不背离本发明的精神下进行各种修饰与变更。 The present invention can also be practiced or applied by other different specific examples, the details of the specification may also, that various changes and modifications without departing from the spirit of the invention based on various concepts and applications.

[0031] 在具体介绍本发明之前,先说明下本发明一种车辆导航系统的最佳路径算法的主要中心思想:通过对A*算法分析可以发现,A*算法在搜索最佳路径的过程中存在着大量的等长路径。 [0031] Before specific description of the present invention, the central idea of ​​the main first described algorithm is optimal path in a navigation system for a vehicle according to the present invention: by analysis can be found in the A * algorithm, A * search algorithm is an optimal path in the process there are a lot of paths of equal length. 本发明中对等长路径的定义是:如果两条路径具有相同的起点和终点,而且可以通过对其中一条路径的组成子路径进行置换重组而得到另一条路径,那么称这两条路径为等长路径。 Definition of the invention is the long path to the other: if two paths have the same start and end points, but also may be obtained through another path wherein the path of a path consisting of sub-replacement recombination, then the other of said two paths long path. 图1中所示的几条路径互为等长路径,通过观察可以得到,每一条路径都是由9个竖直方向和9个水平方向的子·路径组成的。 Several paths shown in FIG. 1 and the like mutually path length can be obtained by observing, each path is the path 9 by the sub-vertical and nine horizontal direction thereof.

[0032] 为了减少A*算法中对等长路径的不必要的搜索和遍历,进一步提高算法的效率和运行速度,本发明提出了一种TPA* (Turning Point)算法。 [0032] In order to reduce the unnecessary traversal search and the A * algorithm in the long path, to further increase the efficiency and speed of the algorithm, the present invention proposes a TPA * (Turning Point) algorithm. TPA*算法的中心思想是删减任何可以从当前节点的父节点以最佳的方式到达的邻接节点,通过这种方式可以避免大量的等长路径。 The central idea is that TPA * algorithm can be cut to any adjacent node from the parent node of the current node in the best way to reach, as long as a large number of paths can be avoided in this way. 另外,在TPA*算法中,所有待处理的节点都是那些在路径前进方向上可能发生方向转折的节点,在两个转折节点之间的部分节点并不进行处理,这就意味着需要遍历的节点数量同A*算法相比会大幅度的减少。 In addition, TPA * algorithm, all nodes are pending that could turn the direction of the nodes occurs in a forward direction path is not processed in the part of the node transition between two nodes, which means the need to traverse the number of nodes will be greatly reduced compared with the a * algorithm.

[0033] 下面结合附图,对本发明的具体实施方式作详细的描述。 [0033] below with the accompanying drawings, specific embodiments of the present invention will be described in detail.

[0034] 步骤一:为了寻找车辆导航系统的最佳路径,首先需要建立本发明算法的具体实施环境。 [0034] Step a: In order to find the best route to the vehicle navigation system, the algorithm of the present invention, first need to establish a specific implementation environment. 本发明一种车辆导航系统的最佳算法的实施是在二维无方向权重相同的网络节点栅格地图中进行的。 BEST algorithm of a vehicle navigation system of the present invention is not entitled to the two-dimensional direction in the same network node weight raster map. 其中,每个网络节点n的邻域节点集合neighbors (n)都含有8个节点,根据车辆行驶地图不同地理特征,每个节点可以有可通行和障碍两种状态。 Wherein each node of the network neighborhood set of neighbors of node n (n) contains eight nodes, depending on the geographic features of the vehicle with the map, each accessible node can have two states and disorders. 一个节点可以通过直线移动和斜线移动两种方式到达另一个节点,相邻节点间的距离默认为I。 Two ways a mobile node can reach another node and oblique linear movement, the distance between adjacent nodes by default I.

[0035] 步骤二:对节点n的邻接节点集合neighbors (n)进行删减,目的是要确定出同到达目标的最佳路径不相关的节点X。 [0035] Step Two: an adjacent node of the node n set of neighbors (n) for the deletion purpose is to determine the best route to the target node with irrelevant X. 为了达到上述要求,需要对两类路径的长度进行比较:^,该路径从节点n的父节点p (n)开始,经由节点n并到达节点x ; ',该路径同样是从节点n的父节点p (n)开始,但是不经过节点n而到达节点X。 To achieve the above requirements, it is necessary for the length of the two paths are compared: ^, the path from the parent node n p (n) starts, via the node to node n and x; ', the same path from the parent node n node p (n) starts, but does not reach the node n through node X. 需要注意的是,上述两条路径中所包含的节点都包含在集合neighbors (n)中。 It should be noted that the above-described two nodes included in the path are included in the set of neighbors (n) of the. 根据父节点p(n)到节点n的方向不同而需要考虑两种情况,分别是直线移动和斜线移动。 The parent node p (n) to the node n different directions we need to consider two cases, namely, linear movement and oblique movement. 另外,如果节点n是起点,因为其父节点为空,故而无法进行邻接节点的删减。 In addition, if the node n is the starting point, because the parent node is empty, and therefore can not be cut adjacent nodes.

[0036] 1、直线移动 [0036] 1, the linear movement

[0037] 在这种情况下,对满足公式(I)中约束条件的任何节点X (XG neighbors (n))进行删减: [0037] In this case, satisfying the formula (I) for the deletion of any constraint node X (XG neighbors (n)):

[0038] Ien (〈p (n),…,x>\n) < Ien (〈p (n),n, x>) (I) [0038] Ien (<p (n), ..., x> \ n) <Ien (<p (n), n, x>) (I)

[0039] 如图2 (a)中所示的情况,p (n) =7,除了节点x=2之外的所有节点都将被删减。 The case shown in (a) of [0039] FIG. 2, p (n) = 7, in addition to all the nodes other than x = 2 will be cut.

[0040] 2、斜线移动 [0040] 2, the mobile slashes

[0041] 同上述直线删减规则相似,但是在斜线移动的情况下,对节点X的删减限制条件更加的严格,需要满足公式(2)中的约束条件: [0041] with the deletion rule is similar to the straight, but in the case of oblique movement, deletion restrictions on node X more strictly necessary to satisfy equation (2) the constraints:

[0042] Ien (〈p (n),…,x>\n)〈len (〈p (n),n, x>) (2) [0042] Ien (<p (n), ..., x> \ n) <len (<p (n), n, x>) (2)

[0043] 如图2 (c)中所示的情况,p(n)=l,除了节点x=5, x=7,x=8之外的所有节点都将被删减。 The case shown in (c) in [0043] FIG. 2, p (n) = l, in addition to the node x = 5, x = 7, all nodes other than x = 8 will be cut.

[0044] 本发明中涉及的算法在上述的两种情况中,neighbors (n)中都没有包含障碍节点,将此时经过删减之后得到的节点称为节点n的普通邻接节点(Normal Neighbor)。 [0044] The present invention relates to a method in the above two cases, neighbors (n) are not included obstacle node, this time after the deletion node obtained in the adjacent node is called common node n (Normal Neighbor) . 而当neighbors (n)中含有障碍节点时,会出现不能完全对非普通邻接节点进行删减的情况,这时称这些节点为特殊节点(Special Neighbor)。 When neighbors (n) contains obstacles node, there will not be entirely the case of non-ordinary adjacent node for exclusion, then call these nodes are special nodes (Special Neighbor). 如果一个节点x (x G neighbors (n))是特殊节点,那么应该满足:节点X不是节点n的普通节点,并且满足公式(3)中的约束条件。 If a node x (x G neighbors (n)) is a special node, should satisfy: X is not a common node of the node n, and satisfies the constraint condition equation (3).

[0045] len «p (n), n, x» <len «p (n), •••, x>\n) (3) [0045] len «p (n), n, x» <len «p (n), •••, x> \ n) (3)

[0046] 图2 (b)中的x=l节点和图2 (d)中的x=3节点都是满足上述条件的特殊节点。 Special nodes [0046] FIG. 2 (b) of the node x = l and 2 (d) x = 3 in all nodes satisfying the above conditions.

[0047] 本发明所涉及的算法在完成了对节点n周围的8邻域节点的删减工作之后会得到一个节点n的邻域节点集合neighbor (n)。 After the algorithm [0047] The present invention relates to the completed cut of the work 8 neighbor nodes around node n to node n will be a set of neighborhood nodes neighbor (n). 考虑在最恶劣的情况下,即节点n是路径的起点nstart时,集合neighbor (n)中的节点数目是8。 When considered in the worst case, i.e., the starting point node n nstart path, set the number of neighbor nodes (n) is 8. 考虑一般情况,并且在节点n的邻域中没有特殊邻接节点(Special Neighbor)的情况下,集合neighbor (n)中的最大节点数量在直线移动时是I,而在斜线移动时是3 ;在存在特殊节点的情况下,集合neighbor (n)中的最大节点数量在直线移动时是3,在斜线移动时是5。 Consider the case where the general, and is not particularly adjacent node (Special Neighbor) in the neighborhood of the node n, the maximum number of nodes in the neighbor set (n) at the time of linear movement is I, and is 3 when moving hatched; in the presence of a particular node, the nodes set the maximum number of neighbor (n) 3 is moved in a straight line, oblique line is 5 when moved. 经过删减之后的邻域节点数量有了显著的下降。 After a number of neighborhood nodes exclusions have been significantly reduced.

[0048] 步骤三:在本发明算法的后续工作中,并不直接将步骤二中得到的邻域节点作为下一步待处理的节点,而是进行对路径转折点的搜索工作。 [0048] Step Three: In the follow-up algorithm of the present invention is not directly neighbor node obtained in step two as a node to be processed next, but instead of turning point route search work. 在此步骤的搜索中,节点n到节AxeneighboHn)的移动方向5是一个需要考虑的关键因素。 In this step of the search, the node n to the section AxeneighboHn) in the movement direction 5 is a key factor to be considered. 本发明所涉及的算法对转折点的定义及搜索示例如下: Algorithm of the present invention is to search and define turning points example:

[0049] 节点m是节点n的后继节点,n到m的方向为t/如果n在此方向上经过最少k次 [0049] m is a successor node of node n, n is m the direction of t / n if happened in this direction through the k

单位移动后到达m,BPm = n + kd,并且满足下列约束条件之一,则称节点m是节点n的转折点。 After the mobile unit reaches m, BPm = n + kd, and one of the following constraint conditions is called node m to node n turning point.

[0050] I•节点m是终点。 [0050] I • m is the end node.

[0051] 2.节点m的邻接节点中至少有一个是特殊节点。 Adjacent node [0051] 2. The node m at least one particular node.

[0052] 3.g是斜线方向时,存在一个节点在^ e彳K}方向上距离m节点Ici个单位距离,并且节点P是满足上述条件I或条件2的一个转折点。 [0052] 3.g is oblique direction, the presence of a node m from node Ici unit distance in the direction K} ^ e left foot, and the node P is the condition satisfying the above conditions, or a turning point I 2.

[0053] 图3 (a)中的节点m是节点n的一个转折点,满足上述条件2。 [0053] FIG node 3 m (a) is a turning point node n, 2 satisfy the above conditions. 图3 (b)中所示的节点m是满足条件3的转折点。 Node 3 shown in FIG. M (b) it is a turning point satisfies condition 3. 从节点n开始沿斜线方向移动,直到遇到节点m,从m开始经过ki=2次水平移动到达节点p,而p点是m点的一个转折点(满足条件2),因此,节点m是节点n的一个转折点。 Begins to move in a diagonal direction from the node n, m until it encounters the node, the elapsed ki = 2 m times the horizontal movement arrives at the node p, the point p and the point m is a turning point (the condition 2), therefore, m is the node a turning point node n.

[0054] 步骤四:将按照步骤三中定义得到的,属于节点n的转折点存放到数组ProcessList中,任意节点XG ProcessList是待处理的节点。 [0054] Step Four: the definition obtained according to step III, the turning point stored in the node n belongs to the ProcessList array, a node is any node XG ProcessList be treated. 在每一次向ProcessList中插入节点的时候,都需要保持ProcessList的有序性,即ProcessList中的第一个节点的路径开销值(从起点到此节点的行进距离)最小,最后一个节点的路径开销值最大。 When every node is inserted into ProcessList, the need to maintain orderly ProcessList, that path cost value of the first node in the ProcessList (travel distance from the starting point to this node) minimum, the last node of the path overhead maximum value. 在下一次算法循环的时候,取出ProcessList中的第一个节点进行处理。 When the next cycle of the algorithm, the ProcessList removed first node for processing.

[0055] 步骤五:循环执行上述步骤二至步骤四,当终点ngMl被插入到ProcessList中时,完成对最佳路径的搜索,算法结束。 [0055] Step Five: loop above steps two to step four, when the end is inserted into the ProcessList ngMl, complete optimal path search algorithm ends. 从终点ngMl开始,依次将其前继节点即父节点添加到路径数组PathList中,直到将起点nstart添加到PathList中。 Beginning from the end ngMl, in turn, following the node that is the parent node before adding it to the array of paths PathList until the start of nstart added to PathList in. 路径数组PathList中保存的即为寻找到的最佳路径上的所有路径节点。 All nodes on the path to save the best path is the path array PathList in looking into. 车辆导航系统可以按照此最佳路径对行驶在地图上的车辆进行实时的导航工作。 Vehicle navigation system can work in real-time navigation for driving a vehicle on a map in accordance with this best path.

[0056] 综上所述,本发明一种车辆导航系统的最佳路径算法在执行的过程中只对地图中的路径转折点进行识别和选取,从一个转折点到其相邻的下一个转折点的中间节点可以被忽略,进而大量地减少了需要处理的节点数量。 [0056] In summary, the optimal path algorithm in a vehicle navigation system according to the present invention, the turning point of the route map to identify and select only the implementation of the process, from the middle of a turning point to the next adjacent inflection point nodes can be ignored, thereby substantially reduced the number of nodes need to be addressed. 本发明提出的算法简单,不需要对地图进行任何的预处理和额外的计算机存储空间开销,可以高效率的搜索出最佳路径,提高车辆导航的效率。 The proposed algorithm of the present invention is simple and does not require any pretreatment map and additional computer memory space overhead, can efficiently search the best route, the vehicle navigation efficiency. 采用本发明,可以很好的在复杂地图环境下进行对车辆路径的规划,迅速为车辆寻找最佳路径。 According to the present invention can be a good vehicle path planning in a complex environment map, quickly find the best path for the vehicle.

[0057] 以上所述仅为本发明的优选实施方式,本发明的保护范围并不仅限于上述实施方式,凡是属于本发明的原理的技术方案均属于本方面的保护范围,对于本领域的技术人员而言,在不脱离本发明的前提下进行的若干改进和润饰,这些改进和润饰也应视为本发明的保护范围。 [0057] The above are only preferred embodiments of the invention, the scope of the present invention is not limited to the above embodiments, all belonging to the principles of the present invention belong to the technical scope of the present embodiment aspect, to those skilled in the art for a number of improvements and modifications made without departing from the present invention in the premise, these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims (1)

1.一种车辆导航系统的最佳路径计算方法,其特征在于包括下列步骤: 步骤一:建立二维无方向、权重相同的网络节点栅格地图,其中,每个网络节点n的邻域节点集合neighbors (n)中都包含有8个节点,根据车辆行驶地图不同地理特征,每个节点有可通行和不可通行两种状态,一个节点通过直线移动和斜线移动两种方式到达另外一个相邻的节点,相邻节点间的距离默认为I ; 步骤二:对节点n的邻接节点集合neighbors (n)进行删减,目的是要确定出同到达目标的最佳路径不相关的节点X,根据父节点P (n)到节点n的方向不同而需要考虑两种情况,分别是直线移动和斜线移动;另外,如果节点n是起点,因为其父节点为空,故而无法进行邻接节点的删减; (1)、直线移动在这种情况下,对满足公式(I)中约束条件的任何节点X (XG neighbors (n))进行删减: Ien (<p (n),…,x>\n)≤ Ien «p Optimal path calculation method for a vehicle navigation system, comprising the following steps: Step 1: establishing two-dimensional non-directional, the same weight map grid network nodes, wherein each network node's neighborhood nodes n a set of neighbors (n) are included eight nodes, depending on the geographic features of the vehicle with the map, and each node has accessible impassable two states, a mobile node through a linear movement and oblique relative to the other two ways to reach adjacent nodes, the distance between adjacent nodes by default I; step two: node n of the set of adjacent node neighbors (n) for the deletion purpose is to determine the best path to reach the same destination node uncorrelated X, the parent node P (n) to the node n different directions need to consider two cases, namely, linear movement and oblique movement; Further, if the node n is the starting point, because its parent node is empty, and therefore can not be adjacent node deletion; (1), the linear movement in this case, any node X satisfies the equation (I), constraints (XG neighbors (n)) for exclusion: Ien (<p (n), ..., x > \ n) ≤ Ien «p (n), n, x>)(I) (2)、斜线移动同上述直线删减规则相似,但是在斜线移动的情况下,对节点X的删减限制条件更加的严格,需要满足公式(2)中的约束条件: Ien (〈p (n),…,x>\n) <len «p (n), n, x» (2) 在上述的两种情况中,neighbors (n)中都没有包含障碍节点,将此时经过删减之后得到的节点称为节点n的普通邻接节点(Normal Neighbor);而当neighbors (n)中含有障碍节点时,会出现不能完全对非普通邻接节点进行删减的情况,这时称这些未被删减掉的节点为特殊节点(Special Neighbor);如果一个节点x (x G neighbors (n))是特殊节点,那么它应该满足:节点X不是节点n的普通节点,并且满足公式(3)中的约束条件。 (N), n, x>) (I) (2), with the oblique movement of the linear deletion rule is similar, but in the case of oblique movement, deletion restrictions on node X more strictly necessary to satisfy constraint equation (2): Ien (<p (n), ..., x> \ n) <len «p (n), n, x» (2) in the above two cases, neighbors (n normal adjacent node (normal Neighbor)) are not included obstacle nodes, this time after the deletion node obtained is called the node n; and when the neighbors (n) containing disorders node, there will not be completely non-common deletion of a case where the adjacent node, when the called node is not cut out of the particular node (special Neighbor); if a node x (x G neighbors (n)) is a particular node, then it should satisfy: node X ordinary node is not the node n, and satisfies the constraint equation (3). Ien (〈p (n), n, x» <len «p (n),…,x>\n) (3) 步骤三:进行对路径转折点的搜索工作,节点n到节点XG neighbor (n)的移动方向;? Ien (<p (n), n, x »<len« p (n), ..., x> \ n) (3) Step three: conduct path inflection point searching operation, node n to node XG neighbor (n) the direction of movement;? 是一个关键因素,对转折点的定义如下: 节点m是节点n的后继节点,n到m的方向为3,如果n在此方向上经过最少k次单位移动后到达= n + 并且满足下列约束条件之一,则称节点m是节点n的转折点; 1) •节点m是终点; 2).节点m的邻接节点中至少有一个是特殊节点; 3).3是斜线方向时,存在一个节点/? = ?« + &式在式£{02}方向上距离m节点ki个单位距离,并且节点P是满足上述条件I或条件2的一个转折点; 步骤四:将按照步骤三中定义得到的、属于节点n的转折点存放到数组ProcessList中,任意节点XG ProcessList是待处理的节点,在每一次向ProcessList中插入节点的时候,都需要保持ProcessList的有序性,即ProcessList中的第一个节点的路径开销值:即从起点到此节点的行进距离最小,最后一个节点的路径开销值最大,在下一次算法循环的时候,取出ProcessList中的第一个节点进行处理 Is a key factor, defined turning points as follows: node m is the successor node n,, n to m in the direction 3 reaches = n if n least k times the unit is moved in this direction through the + and the following constraint conditions one, called the node m is the turning point node n; 1) • m is the end node;. 2) in the adjacent node m is at least one particular node; 3) .3 is a diagonal direction, the presence of a node ?? / = «+ formula in formula £ {02} from the direction of the m nodes ki units of distance, and the node P satisfying the above condition I or condition a turning point 2; step four: obtained in step defines three in accordance with the turning point stored in the node n belongs to the ProcessList array, a node is any node XG ProcessList be treated, when each node is inserted into the ProcessList, are required to maintain orderly ProcessList, i.e., the first node of the ProcessList path costs: the minimum travel distance from the start point to this node, the last node of the maximum path costs, when the next cycle of the algorithm, the ProcessList taken for processing the first node 步骤五:循环执行上述步骤二至步骤四,当终点ngMl被插入到ProcessList中时,完成对最佳路径的搜索,结束;从终点ngMl开始,依次将其前继节点即父节点添加到路径数组PathList中,直到将起点nstart添加到PathList中,路径数组PathList中保存的即为寻找到的最佳路径上的所有路径节点,车辆导航系统可以按照此最佳路径对行驶在地图上的车辆进行实时的导航工作。 Step Five: loop above steps two to step four, when the end is inserted into the ProcessList ngml complete the search for the best path, ending; ngml starting from the end, i.e. predecessor node sequentially added to the parent node array of paths PathList until the start of nstart add to PathList, all nodes on the path to find the best path is the path to save the array PathList, the vehicle navigation system in real time on the map for driving a vehicle in accordance with this best path navigation work. · ·
CN201310297623.5A 2013-07-16 2013-07-16 Optimal path calculation method for vehicle navigation system CN103344248B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310297623.5A CN103344248B (en) 2013-07-16 2013-07-16 Optimal path calculation method for vehicle navigation system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310297623.5A CN103344248B (en) 2013-07-16 2013-07-16 Optimal path calculation method for vehicle navigation system

Publications (2)

Publication Number Publication Date
CN103344248A true CN103344248A (en) 2013-10-09
CN103344248B CN103344248B (en) 2015-07-08

Family

ID=49279064

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310297623.5A CN103344248B (en) 2013-07-16 2013-07-16 Optimal path calculation method for vehicle navigation system

Country Status (1)

Country Link
CN (1) CN103344248B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104464280A (en) * 2014-09-05 2015-03-25 广州市香港科大霍英东研究院 Vehicle advance expenditure prediction method and system
CN104717278A (en) * 2014-12-04 2015-06-17 广西蕴藏科技有限公司 Mark position guiding method and system
CN104917835A (en) * 2015-06-10 2015-09-16 广西蕴藏科技有限公司 Mark position guidance system of small network traffic and guidance method therefor
CN105606113A (en) * 2016-01-28 2016-05-25 福州华鹰重工机械有限公司 Method and device for rapidly planning optimal path
CN105739495A (en) * 2016-01-29 2016-07-06 大连楼兰科技股份有限公司 Driving path planning method and device and automatic steering system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1085499A2 (en) * 1999-09-17 2001-03-21 Philips Corporate Intellectual Property GmbH Spelled mode speech recognition
US6259988B1 (en) * 1998-07-20 2001-07-10 Lockheed Martin Corporation Real-time mission adaptable route planner
CN1938562A (en) * 2004-03-31 2007-03-28 英国电讯有限公司 Pathfinding system
CN101900570A (en) * 2009-02-18 2010-12-01 三星电子株式会社 Apparatus and method for generating and using a grid map path
CN102521328A (en) * 2011-12-06 2012-06-27 上海京颐信息科技有限公司 Optimization method for track playback function in indoor positioning system
CN102778229A (en) * 2012-05-31 2012-11-14 重庆邮电大学 Mobile Agent path planning method based on improved ant colony algorithm under unknown environment
CN102901500A (en) * 2012-09-17 2013-01-30 西安电子科技大学 Aircraft optimal path determination method based on mixed probability A star and agent

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6259988B1 (en) * 1998-07-20 2001-07-10 Lockheed Martin Corporation Real-time mission adaptable route planner
EP1085499A2 (en) * 1999-09-17 2001-03-21 Philips Corporate Intellectual Property GmbH Spelled mode speech recognition
CN1938562A (en) * 2004-03-31 2007-03-28 英国电讯有限公司 Pathfinding system
CN101900570A (en) * 2009-02-18 2010-12-01 三星电子株式会社 Apparatus and method for generating and using a grid map path
CN102521328A (en) * 2011-12-06 2012-06-27 上海京颐信息科技有限公司 Optimization method for track playback function in indoor positioning system
CN102778229A (en) * 2012-05-31 2012-11-14 重庆邮电大学 Mobile Agent path planning method based on improved ant colony algorithm under unknown environment
CN102901500A (en) * 2012-09-17 2013-01-30 西安电子科技大学 Aircraft optimal path determination method based on mixed probability A star and agent

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104464280A (en) * 2014-09-05 2015-03-25 广州市香港科大霍英东研究院 Vehicle advance expenditure prediction method and system
CN104717278A (en) * 2014-12-04 2015-06-17 广西蕴藏科技有限公司 Mark position guiding method and system
CN104917835A (en) * 2015-06-10 2015-09-16 广西蕴藏科技有限公司 Mark position guidance system of small network traffic and guidance method therefor
CN105606113A (en) * 2016-01-28 2016-05-25 福州华鹰重工机械有限公司 Method and device for rapidly planning optimal path
CN105739495A (en) * 2016-01-29 2016-07-06 大连楼兰科技股份有限公司 Driving path planning method and device and automatic steering system
CN105739495B (en) * 2016-01-29 2019-05-14 大连楼兰科技股份有限公司 Planning driving path planing method, device and automatic steering system

Also Published As

Publication number Publication date
CN103344248B (en) 2015-07-08

Similar Documents

Publication Publication Date Title
Silver Cooperative Pathfinding.
Hilger et al. Fast point-to-point shortest path computations with arc-flags
Jing et al. Hierarchical optimization of optimal path finding for transportation applications
US20140032113A1 (en) Automated route determination
Ferguson et al. Field D*: An interpolation-based path planner and replanner
Huang et al. Large scale real-time ridesharing with service guarantee on road networks
Wang et al. Fast and Memory-Efficient Multi-Agent Pathfinding.
Rousseau et al. Interior point stabilization for column generation
Arslan et al. Use of relaxation methods in sampling-based algorithms for optimal motion planning
CN103593430B (en) A segmented target temporal clustering information based on the movement trajectory method
Chen et al. Reliable shortest path finding in stochastic networks with spatial correlated link travel times
US20130060468A1 (en) Journey planning in public transportation networks
CN103134505B (en) Path planning system and method
CN101241507A (en) Map road-seeking method and system
Fraigniaud et al. Collective tree exploration
CN101944095B (en) Path planning method and system
JP2003233768A (en) Dual dijkstra&#39;s algorithm for searching a plurality of routes
US20120310523A1 (en) Customizable route planning
CN103324982B (en) A path planning approach based on genetic algorithm
CN1908588A (en) Fast map matching method based on small lattice road network organization and structure
Delling et al. Engineering time-expanded graphs for faster timetable information
CN101493518A (en) Wireless sensor network node positioning method and device
CN102679990A (en) Navigation method and device
US20020169543A1 (en) Long distance routing
Cao et al. Sharek: A scalable dynamic ride sharing system

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model
CF01 Termination of patent right due to non-payment of annual fee