TW201042238A - System for providing multiple paths applying to a road map and method there of - Google Patents

Abstract

A method of providing multiple paths is applying to a road map, and the method comprising the steps of: defining an initial node and a terminal node in the road map. Sorting the path costs which between the initial node and the neighbor nodes thereof, and then storing the path costs in an analysis array. Further, search the following nodes of initial node through the neighbor nodes in order, and calculate separately the path costs of the initial node to the following nodes, and accumulates to store in the analysis array until the first path cost of final node is greater than or equal to the last path cost of terminal node. Wherein, the path costs store in the different areas of analysis array according to the differentiation of nodes. Hence, the present invention can achieve the purpose of providing the multiple paths more efficiently.

Description

201042238, invention description: [Technical Field] The present invention relates to a driving route planning system and method thereof, and a driving pure planning system and method thereof for providing a multi-record short path. [Prior Art] Since the graph network theory (G_ and Net_k(10)(4) is a theoretical science in mathematics application, the _ algorithm is two kinds of methods that can be directly given (four), which can directly explain it. The purpose and the way of performing actions are related. The theory of the network in the daily life is also very extensive. For example: pure network configuration and production: distribution problems, cargo transportation and distribution system, urban transportation system, Wang Qiong Position systems, facility layout, file consolidation systems, routes and scheduling systems, circuit telecommunications, and even water f distribution lines can be solved by the theory of the map network to save _, cost and increase convenience. In terms of the problems solved by the graphical algorithm, most of them are still to solve the problem of the shortest path (Sh_st Path). At present, the shortest path of the conventional method is roughly: Qi Qianlin (Breadth)

First S_h, BFS), Dexter's algorithm (Dijk, s Mgodthm), decentralized routing algorithm (Bdiman F〇rd

Alg〇rithm), Freudian algorithm (10), foreign law marriage and A* search method. Taking the Dexter algorithm as an example, the main spirit is to use a certain point as the starting point, and in the $ point that is connected to the starting point but has not been selected, the node with the shortest distance from the starting point is selected. And ride through the nodes of the ride to update the other festival distances. For example, 201042238 is "so far", that is, the first picture: two, the operation process diagram of the implementation algorithm: and "edge", and each "edge" and the two nodes belonging to it __, example; ^ has its distance Value to represent the distance between the b nodes. And (a, b) refers to the result of the a node and the step of filling in the steps. The ^ table is based on the algorithm Ο — step -: assuming a node as a node. 'The mouth is added to the new node step one. In the node that is added to the a section is not force- has b node and === two =, = off = and: the step of the scheduled tender - after adding the d node, the graphics In the case where the node is not added, the d is connected to the d node, and the right is ''. =9, λ!, the distance value of the present, so 69 is less than 3+2=5 in step 2, d ? E (^> The distance is shorter, thus == f 5 and due to ·) Step 4: After force 2 ==_ is ° 八c is the point, the node in the graph gives = input node has b node. Therefore, in the second transmission: = sub heart 'E (a, _ distance is 5 + 2 b 26, greater than The situation of ^ in step 3 then retains the original value of the distance of 9, the b-node, so the node that has not joined the b: is only after the b-point is added to the algorithm. Thus, the score is 5 201042238 The shortest distance between the b, C, and d nodes and the a node can be obtained. = As can be seen from the above, the main purpose of the shortest path algorithm is to provide the shortest path of the two nodes, and this is also the difference between the two nodes. Between For other alternative paths, in ί, although the / day ^ can respond to (4) the field will be relatively limited. This short circuit / ^ conventional algorithm can smoothly find the most between the two nodes "test = master Calculate the efficiency of the search, so it can not be effective in the application of the time-to-time efficiency application. [Inventive content] Point accumulates the stored path weights, and then ten pairs of different sub-section paths. Therefore, not only can the shortness of j be satisfied to be used for selection = so that the other alternatives are used after the other sorting of the ancients. Therefore, it is convenient to respond to various application fields ^ In order to achieve the above purpose, according to the weight The party: point = the metric in the f between the nodes - the starting node and one: the first. In the first subnode of the subsection, the following: P,, and get with the starting The node diameter weights are sorted and performed: save: 乂: the path J between the adjacent child nodes stores the minimum route avoidance secondary, and the obtained child node obtained by the analysis matrix is set as _! Child node, and will relay the point, then perform a search with the t-segment 201042238 The sub-nodes of the successive nodes are calculated by the weights of the starting nodes, and the successive sub-nodes are counted to the starting, and each area is updated and stored in the corresponding regional node of the analysis array, and then sorted by two or two. Excluding the obtained sub-weighted area represented by 'analysis, where the minimum path is stored in the column, and the node will continue to be connected' to calculate the starting node to the middle of the array in the analysis, that is, between the points The path weights until the first path weight in the area of the area 17 equal to the terminating node is greater than the back path weight in the area of the terminating node. Then the column is ==: algorithm to _ multiple paths The case provided by nt according to the present invention includes: a path planning system for providing multiple paths, wherein the wheel unit is a single database unit and a network analysis unit. To store a plurality of child nodes, sections;:: wide data, and the database unit uses the shape and the sub-sections of the two sub-sections to form at least a road network diagram 〇 is the path weight of the connected round-in unit. The road network analysis unit corresponds to the road network graphics and carries on more: single-input and: = round the data to obtain the weight, you have two multiple re-borrowing roads: 'According to the calculated path power into the overview and the following The detailed description and the purpose of the present invention are for the purpose of achieving the intended purpose, the traitor and other purposes, and (4), which will be followed by the second embodiment of the present invention. The method of multipathing is real 7 201042238 ^ column flow chart. As shown in the figure, the present invention provides that multiple paths are applied to the road network graphics, and the road network maps are the points and the points represented by the child nodes. The method comprises the steps of: first, in the child nodes, a point j and a terminating node 'and obtaining a point (S2G1) with the starting node. Then, referring to the road network graphic, according to the starting point Sort and save

Different areas in the middle. In addition, the length y 设定 set by the area of the analysis array is analyzed, and the mouth is set according to the number of paths to be provided finally. Further, after the step (S2G3), the child node represented by the region in which the subroutine point is stored in the analysis array can be obtained, and the obtained relay node (s2〇5) is further obtained. Then, the search relay node (S207) is further searched and the path weight of the relay node 3= to the start node is further calculated to update the storage location; (_) in the corresponding region of the array. Wherein, after the exclusion, the road can be controlled, so in the step (S209), the path weight is calculated, t is based on the connection path between the different sub-i of the existing path weight in the analysis_ to further calculate the connection. The sub-weight of the relay node: μ advance: the path weight of the child node to the start node is connected. The above-mentioned update storage in the analysis array is performed by stacking to store the path weight. Eight, after the step (s2G9), the small and large path weights are further sorted (s2u) for each region in the analysis array. And it is also mentioned that after the sorting from small to large, if the calculated weight from the number of the area of the analysis array is exceeded, 201042238 does not proceed, and the path exceeding the length of the sub-region is stored. Weights. Then ^ί者' is judged to analyze whether the last subsection of the array represents a path weight greater than or equal to the final 4; the last path weight in the 3 domain). Among them, the termination of the stomach: the last path weight in the & field is · ', the area of the point has been filled and there is actual; when the preset value is stored, the judgment of the step can be established as = Ο Ο = If the result of the step (S213) is YES, it means that the termination is *, the area has filled the path weight, and the judgment result is A曰#, that is, two: rrr is greater than or equal: stop=two: - : Sub-section 2: The end of the festival can be selected: (S2L all roads correspond to the path of the sub-segment 2_213), the result of the touch, do not escape the last - section. ====: / small - before The filling of the path weight is terminated, so the heart of the node is currently terminated = two: = weight is taken, and the value of the wireless is preset. In the above-mentioned execution step 5), the child node (S217)' of ^ is obtained and the sub-section 9 201042238 represented by the region of the weight is repeated until the judgment result of the step (S213) is YES, and the column 2 = By using the present invention, it is possible to provide multiple short-distance followers ί: ΓΛ Illustrate the method of the present invention for providing multiple paths, please re-intention. It has five sub-nodes (A;, c= mention, road network graphics, path weights between, for example, to the table: to provide two more than the length of each region in this embodiment is the two analysis array 1 The process is defined as follows; first, in the road network graph, then:: Zhao Di: ί: = ' and define the child node D as the termination node. The neighbor child node (B: C) diagram: = _^ The weight is in the different regions of the path E(A, C). For example, in the case of the first: 析 歹 quot , , , 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存 储存: is the area of child node C, because there are two: the area of the weight weight

Click to search. In addition - add c into the picture: 3 points C into the technology, after the child node C will be two C, it is known as the connection child node (B, D and E two son: f ' can be conditionally under the road network graphics, Calculate the path weight of the starting node through sub-node C to accumulate 201042238 11; and E(A, D)=2.2+2.5=4 7 1 , /*匕1, 'domain 12; and E(A, E)=2.2+9,9:112^Adding the area 13 stored in the sub-node D. Among them, the accumulated path weights are further stored in the area 11 of the cumulatively stored two nodes E. The sub-node D has a region 12 and a sub-, ,,: sorting 'and the new storage will also pass through the small to large row 13 area, and at the same time, the row can be added A sub-node after storage. E is the most embarrassing in the road network graph. So, immediately after the _ _ 胄 个 路径 路径 路径 路径 路径 路径 路径 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否 是否It is not filled up, so the area 12 of the sub-D area D is still pre-emptive: the result of the judgment is no. After the node, the current analysis (4) is obtained, and the sub-region obtained by the middle-aged generation is replaced. For the child nodes of the table, please refer to the third map for the weight of the two paths, and set the sub-node B as the relay dumper for 11 Π to search for the sub-nodes that the sub-nodes will continue to connect. For the child nodes (C and D), then the ^ is the point and the road network graph, the calculation is wonderful _ Α points; the condition of the point 重 is heavy, E (A, C) = 3.1 + 13 = 44 to g (and D) the path weight 1〇)it g. Cumulatively stored in the area of the child node C and the order after the order is unchanged; to be accumulated in the child nodes D h(A, D)~3.1+7.4-10 · 5 order unchanged. Μ 12 'and the sorted after the score Hi can be passed after the exclusion of the path 'Therefore, the path weight already existing in the array 1 must be analyzed to further calculate the node 11 201042238 node can continue to follow This joins child node B to find the weight. That is, after adding the search, and after the child node ^ = child node C has been previously unique to the child node D analysis array i this please refer to the third E picture, in the data content of the join ^ E. When it is necessary to simultaneously consider the sub-node B ^ point B to search, the conditions of the starting node A to: point C must be counted to calculate E (A deciding sub-path weight, ie 12 ' and after sorting) Area of the road

St This embodiment only needs to provide two two: two jobs will not be stored. The towel E(AE)=3 仏榷10.5 is stored in the area 13 of the point E, and is stored after the sorting and when the child node B joins the search: ^: ': D (terminates the point) area The last one in 12: The area 12 is filled, so d Μ replaces the original preset value (infinite illusion. Therefore, the first path weight 121 in the quotient is greater than or equal to the parent node d & field 12 The last path weight in the 69, thus indicating the search for two shorter paths from the starting node A to the terminating node D (the shortest and second shortest paths respectively). # this, you can list the child nodes d The path corresponding to all the path weights in the region ,, that is, the eight 欠 〇 路径 (the path weight 4.7) and the A4B + C + D (the path weight 6.9) ^ and the implementation of the present invention provides a multi-path method for the path planning The system's resource dispatch system, intelligent transportation system and navigation system, which can be applied, for example, to disaster relief strains, requires fast calculations 12 201042238 and has alternative roads (4) for selection systems. Method, heart two; 秘 secret in the Tudor function to achieve the whole The establishment of the system. The application of the additional real two: square: m four map 'is a block diagram of the driving route planning system of the present invention. As shown in the figure, the source of the road routing rules 'network 1 2 its input ordinary 90, - database unit 91 and - road analysis (four) 92 ' and the network analysis unit % step-by-step includes an eight-array array 921. Among them, the input unit 9 is a knife

The database unit 91 stores a plurality of child nodes and a two-way network: the analysis unit % is a connection input unit _ bucket library = to obtain the corresponding road shape according to the input data and update, and further: Referring to the method step J described in the foregoing second figure, the path weights from the starting node to the terminating node and the heap fingers are stored in the analysis array 921+, and sorted by small to large, and the array is analyzed. The "2i X-step is divided into a plurality of regions to store the path weights of the starting node to the different child nodes respectively. Finally, the road network analyzing unit 92 can be based on the path weights in the region where the node is terminated. In addition, the driving path gauge (4) of the present invention is further distributed by the source of disaster relief, and it is assumed that there are rural disaster relief equipment stored in multiple warehouses, and each warehouse In the number of disaster relief equipment stored in the area, when a disaster occurs in a certain place to form a demand site, the appropriate source of poverty must be dispatched from each warehouse according to the demand of the H material. 9〇 is the information used to provide input disasters to generate information about the people, such as the equipment used to input the current disaster situation, the location of the disaster 13 201042238 and the location of the disaster situation / 3⁄4 s - is stored in the local warehouse (subg = Beku early in the 91 _ (sub-point) and the number of disaster relief equipment in her warehouse __, and the warehouses around the country =, road network knife analysis 兀 92 * in the connection (10) 2 related After the information, 'the second supply of the read database unit 91: = the child node of the place of occurrence 】) The warehouse forms the disaster prevention with the original disaster prevention f, the defect (4) + t, the network analysis unit 92 and the disaster prevention The warehouse is set as the starting point for efficient handling. When the path to the location where the road is planned is blocked or interrupted, the alternative path is selected and the rescue effect is achieved. horse

C = As described above, the method for providing multiple paths of the present invention is to cumulatively store the sorted paths at the same time as the nose path weights, and then to correspond to the path between the starting node and the terminating node. In this way, 堇 can calculate the shortest path in a way that is efficient to read. #Use the same day ί to generate a shorter alternative path after other sorts to choose from, so that the month is convenient for various application fields. It is clear that the multipath method has an operation time of approximately (—(:)) and its dry η is the number of child nodes. The above description is only a detailed description of the specific embodiments of the present invention and is not intended to limit the invention. All of the scopes of the present invention 14 201042238 shall be based on the following claims. Those skilled in the art can readily adapt to variations or modifications within the scope of the invention as defined in the following patents. [Simple diagram of the diagram], a diagram is a schematic diagram of the operation process of the conventional algorithm, and the second diagram is a flowchart of an embodiment of the multi-faceted method; the second diagram is the road of the invention Schematic diagram of an embodiment of a network graphic and its operation process; and a block diagram of an embodiment of the system

The fourth figure is the driving path rule of the present invention [Description of main component symbols] Analysis array 1 Sub-C area 1 Sub-node B area 11 Sub-point D area 12 Sub-node E area 13 Driving path planning system 9 input Unit 90 Database Unit 91 Road Network Analysis Unit 92 Analysis Array 921 15

Claims (1)

201042238 VII. Scope of application for patents: System, the steps of the method =, 祠 祠 拼 的 的 的 的 , , , , , , ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ (b) of the nodes (b) in the child nodes (:=node one and -terminating node' and reordering:: and the eight: child W (d=analysis matrix_ground array) has the smallest path weight stored in the array The sub-(4) of the region =, and the obtained sub-node is set to - relay section ==:r:;r node; ===== : the road is greater than or equal to the termination node curb weight; and (m All the path weights in the area of the terminating node are corresponding to the road provided by the driving path planning system, and the road provided by the driving path planning system is provided as described in the third paragraph of the patent application. The path planning system is provided for the rescue of the two systems, such as the rescue system, the intelligent transportation system and the navigation system, as described in the patent application scope item, wherein the database unit is presetly stored, Control method, the input data generated by the human unit? And the loss, as described in item i of the patent application scope: ^ point of the data. The method of analyzing the length of the area of the array 'provides the number to set. And ', according to the multi-path of the side, such as the scope of patent application The method of providing multiple paths in the area of the analysis array, the method of providing the multipath according to the fifth item of the last section «repetition node of the re-pointing node, _: twenty: the path weight of the output When the area where the terminating node is filled is stored, the method of providing multiple paths as described in the last term of the terminating node (4) can be replaced by the method of the last node (4), which is further based on the analysis array. The existing path weight is 'to calculate the path weight of the child node to which the succeeding child node can further perform to the starting node. For example, the method for providing multiple _ as described in Item 1 is applied," 'The ranking of each region is based on the weight of the path from small to large. As described in the scope of the patent application, the method of providing multiple paths, in which step (e) towel, age analysis (four) more bribes are adopted Stack 2〇1〇4 2238 ίο, way to store. Including the vehicle road control planning system, is used to provide multiple paths, the package is a single:: is used to generate - wheeled data; =, library early (, is stored a plurality of child nodes And at least one network pattern formed by the plurality of sub-networks and a path weight of the whole station; and an I-network analysis unit represented by the points between the I and the sub-points is connected to the input port to obtain the pair of libraries according to the input data The second bank is updated in a row, and then the road network graph is updated: and the % node and the terminating node reach the path weight of the terminating node/the difference is from the starting node of the 4th node. Corresponding to ~ 11, as outlined in the 1G item, the driving route planning system is a salvage system, = intelligent transportation system and navigation system, etc., 12, such as the middle of the special The sub-nodes of 13 and :c described in item () are: planning system, 14 which is in the planning department, and, as mentioned in item 10 of the patent application scope: The road network analysis unit is stepped into the planning system, and the array is stored by the analysis array to store the roads: the array, and the large path weights are used for sorting. 4 heavy, and according to the driving path planning system described in the small patent application scope, 18 201042238 wherein the analysis array is divided into a plurality of regions to store the path weights of the starting node to the child nodes respectively . 19