KR100791748B1 - Reduceing Method of Shortest Path Searching Area and Calculating Method of Minimal Expecting Load and Method of Searching Shortest Path - Google Patents

Abstract

The present invention relates to a method for limiting a search area, a method for obtaining a minimum expected value, and a method for searching the shortest path in a complex network in a faster time. The present invention relates to a method of searching for the shortest path using the minimum circle unit of the spatial distance reference actual required value in the reduced search area.

Shortest path, minimum expected value, clearance

Description

Reducing Method of Shortest Path Searching Area and Calculating Method of Minimal Expecting Load and Method of Searching Shortest Path}

1A and 1B are flowcharts illustrating an embodiment of a shortest path search method using a minimum distance unit based on a distance distance reference time distance according to the present invention;

2 is a flowchart illustrating an embodiment of a method for constructing a temporary shortest path for minimizing a spatial distance according to the present invention.

3 is a diagram illustrating a virtual network for explaining the present invention.

The present invention relates to a search area limiting method for searching the shortest path in a complex network at a faster time, a method for obtaining a minimum expected value using the same, and a shortest distance searching method. The present invention relates to a method for narrowing a search area and searching for a shortest path using a minimum circle unit based on a spatial distance reference actual value in the reduced search area.

The shortest path search is used in various fields such as navigation, communication network, game software, and artificial intelligence. Among the existing search algorithms, typical algorithms are Dijkstra algorithm, Freud algorithm, and A * algorithm. .

Dijkstra's algorithm selects paths that start from the starting node with the minimum time distance of the directly connected path irrespective of the direction to the target node. By comparison, when the network size is very large, the search time becomes considerably longer. There is.

The Freud algorithm also has a problem that the search time of the Dijkstra algorithm is long in that it finds the shortest path by checking and comparing all possible links from the starting node to the target node.

The A * algorithm is representative of the heuristic algorithm and estimates the shortest path using the minimum estimate of the remaining path cost from the current node to the target node. In other words, since the shortest path is searched by using the actual distance from the starting node to the current node and the estimated distance from the target node (typically, the linear distance), the existing Dijkstra algorithm or Freud algorithm are used. Compared with this, the search time is shortened. However, it is difficult to select the minimum estimate of the remaining path cost, and if it is wrongly selected, it may cause no solution. Also, when the estimate is used with a simple straight line, it is inefficient depending on the structure of the network. The search may be performed.

In addition, the conventional search method is performed in an inefficient way to provide the shortest path through the reverse search (recovery) function from the target node to the starting node after the search is completed, using the connection relationship between the nodes again.

An object of the present invention for solving the above problems is to quickly search the shortest path from the starting node to the target node in any network structure, the nodes that are directly connected while minimizing the sum of the spatial distance between the starting node and the target node Establish the temporary shortest path consisting of the two paths, calculate the minimum required unit of the spatial distance based on the distance distance divided by the actual distance such as cost, time and actual distance, and calculate the actual required value of the temporary shortest path based on the spatial distance. The search area is reduced to a range that has a smaller spatial distance than the minimum circle unit, and the minimum required value is calculated using the minimum unit and the spatial distance within the reduced search area. To search the shortest path correctly without searching the entire node while comparing To provide.

A feature of the present invention for achieving the above object is to find the minimum required unit, such as space distance reference time, cost, actual distance, etc. on each link for the shortest path search entire area (total network), starting node and target Construct the temporary shortest path that minimizes the spatial distance while directly connecting to the node, calculate the actual required value of the temporary shortest path, and apply the minimum unit of the spatial distance reference actual required value to the actual required value of the temporary shortest path. Node).

In addition, another feature of the present invention is to obtain a minimum distance unit of the required distance reference based spatial distance, and to select a node that is directly connected to the last search node among nodes connected from the departure node to the arrival node and is not included in the search completion node set and the temporary shortest path. The method is to find a minimum expected value that calculates a minimum expected value by adding the actual required value to the search node and the spatial distance from the search node to the arrival node multiplied by the minimum required unit of the spatial distance reference. .

Another feature of the present invention is to find the minimum unit of distance required based on the distance distance in the reduced search area or the entire network as described above, calculate the minimum expected requirement using the found minimum unit and the space distance A method of searching for the shortest path by repeatedly comparing the required value with the actual required value in the temporary shortest path.

Hereinafter, the shortest path searching method using the search area limiting method according to the present invention will be described in detail with reference to the accompanying drawings.
The following embodiment is implemented through a conventional computer system, navigation, etc., the configuration of the command input unit, such as a keyboard, mouse, touch pad, etc. to select the start / destination node, and a plurality of nodes And a calculation unit for finding the shortest path of the destination node from the departure node by using the memory information according to the input operation control signal, and the configuration of such a system is already known to the supplier. Since a detailed description thereof is omitted, the following embodiment relates to a process in which the calculation unit refers to the contents of the memory after setting the starting node and the target node through the command input unit. Omitted below.

1 is a flowchart illustrating an embodiment of a method for searching a shortest path using a minimum circle unit based on a spatial distance. In an embodiment of the present invention, a required value uses a time distance that is actually required time.

<Step 101>

The entire nodes p1, p2, ..., pn in the search range are included in the set M.

<Step 102>

The minimum value of the time distance divided by the spatial distance is selected for all connected links, and this minimum value is defined as the minimum distance unit (Min_Unit) based on the distance distance for limiting the search area.

The time distance between the pi node and the pj node is expressed as rdis (pi, pj), and the space distance is expressed as sdis (pi, pj). The spatial distance means a linear distance, and can be calculated using the following Equation 1 from the relative coordinates or the absolute coordinates of pi and pj.

In Equation 1, the coordinates of pi are (Xi, Yi) and the coordinates of pj are called (Xj, Yj).

Accordingly, the minimum distance unit based on the distance distance Min_Unit is determined by selecting a minimum value of rdis (pi, pj) / sdis (pi, pj) values in the entire paths.

In the present invention, since the raw unit is used, not only the shortest path of the time distance, but also the shortest path of the actual distance, the shortest path required, and the like can be found.

<Step 103>

Let the starting node be sp and the target node ep.

<Step 104>

As the sum of the space distance from the starting node sp and the space distance to the target node ep is minimized, a temporary shortest path is constructed from nodes directly connected to each other, and a time distance MINP_rdis of the temporary shortest path is calculated.

The calculation process is shown in detail in FIG. 2, in which step (i) is set to '0', the starting node sp is set as the selection node p [0], and stored in the temporary shortest path Tem_MINP [i]. The time distance MINP_rdis in the temporary shortest path is set to '0' (201).

Increment step (i) by '1' (202), directly connected to the previous selection node (p [i-1]) and not included in the temporary shortest path (Tem_MINP [k], k = 0 to i-1). In step 203, a search node p [i] is selected in which the sum of the spatial distance at the departure node sp and the spatial distance at the arrival node ep is minimum.

Save the selected discovery node to the temporary shortest path (Tem_MINP [i]), and store the time distance from the previous discovery node (p [i-1]) to the current discovery node (p [i]). In addition, it is changed (204).

Search for whether the search node p [i] is the target node ep (205). If the target node ep is the target node (ep), the time distance of the temporary shortest path is output (206), and this is referred to as the shortest time distance MINP_rdis. . If the target node ep is not, the process is repeated from step 202.

<Step 105>

The limit spatial distance Lim_sdis for the reduction of the search area may be calculated by dividing the time distance MINP_rdis of the temporary shortest path by the minimum distance unit Min_Unit based on the spatial distance.

That is, a node whose sum of the spatial distance at the starting node sp and the spatial distance at the target node ep is larger than the limit spatial distance Lim_sdis is unlikely to be included in the shortest path of time distance shorter than the temporary shortest path. .

<Step 106>

The search region set NM is reduced based on the starting node sp and the target node ep. The nodes included in the set NM are np1, np2,... npi, the node npi included in the set NM is limited by the sum of the spatial distance sdis (sp, npi) at the departure node sp and the spatial distance sdis (npi, ep) at the arrival node. Nodes smaller than the spatial distance (Lim_sdis).

<Step 107>

For the links in the reduced search area set NM, the minimum distance unit NMIN_Unit based on the distance distance is obtained again as in <Step 102>.

 The target node (ep) at the starting node (sp) has a reduced search target node set (NM) obtained in the above process, the minimum distance unit (NMIN_Unit) based on the spatial distance, and the shortest time distance (MINP_rdis) calculated from the temporary shortest path. The shortest path search up to) is performed.

<Step 108>

The progress step (i) is set to '0', the time distance (c [i]) to the progress step (i) is set to '0', and the starting node is found in the node set FN [i] which has been searched in step ⅰ. Save (sp). The start node sp is also assigned to the search node np [i] and the temporary shortest path node Tem_MINP [i].

<Step 109>

Increment the step by '1' (ⅰ <= ⅰ + 1).

<Step 110>

Nodes connected directly to the last search node (np [i-1]) and not included in the search complete node set (FN [i]) and temporary shortest path nodes (Tem_MINP [k], k = 0 to i-1). And find the search node (np [i]).

<Step 111>

If no matching search node (np [i]) exists, go to <Step 117>.

<Step 112>

The searched node np [i] is stored in the search completed node set FN [i], and the time distance rdis from the previous search node np [i-1] to the current search node np [i]. (np [i-1], np [i]) is added to the time distance c [i-1] from the departure node to the previous search node (np [i-1]) and the time distance c [i]. Modify

The minimum expected time distance c_dis is the minimum expected distance from the current progress stage (i) to the target node, and the target node (ep) at the search node np [i] at the modified time distance c [i]. Calculated by adding the distance distance (sdis (np [i], ep)) up to) times the minimum distance unit (NMIN_Unit).

c_dis = c [i] + sdis (np [i], ep) × NMIN_Unit

Where c_dis is the minimum expected time distance, c [i] is the constructed time distance from the departure node sp to the second-level search node np [i], and sdis (np [i], ep) is the second-level search Space distance from node to target node (ep), NMIN_Unit is minimum distance unit based on space distance.

<Step 113>

If the minimum expected time distance c_dis is greater than the shortest time distance MINP_rdis, go to <Step 110>.

<Step 114>

If the minimum expected time distance c_dis is smaller than the shortest time distance MINP_rdis, the search node np [i] is stored in the temporary shortest path node Tem_MINP [i].

<Step 115>

In step 114, if the search node np [i] is not the target node ep, the process goes to step 109.

<Step 116>

If the search node np [i] is the target node ep in step <114>, the time distance c [i] to the current step is set as the shortest time distance MINP_rdis, and the current step (i) is the shortest path. After the final step (Fi) of, the temporary shortest path (Tem_MINP [i]) from step 0 to the current step (Fi) is converted to the shortest path (MINP [i]) and goes to <Step 110>.

<Step 117>

Remove all nodes in the search completed node set FN [i] in the current step (i) (FN [i] = {}), and decrease the step by one (ⅰ <= ⅰ-1). .

<Step 118>

If step (i) is not '0', go to <Step 110>.

<Step 119>

Output the shortest set of paths (MINP [k], k = 0 ~ Fi) and output the shortest time distance (MINP_rdis).

If the time required for the route is given according to the change period (10 minutes, 5 minutes, etc.), repeat from <Step 103> within the same time change period, and repeat from <Step 102> if it exceeds the change period. You can navigate the path dynamically and efficiently.

Example

3 shows a virtual network, and shows a node number, a connection relation between nodes, and a time distance.

In the virtual network of FIG. 3, the node set M of the search area is as follows (Step 101).

M = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}

The spatial distance calculated from the relative coordinates of the nodes in the virtual network is shown in Table 1 below.

The time distance / space distance values of the directly connected path are shown in Table 2 below.

The minimum value of the time distance / spatial distance value is 0.8929 of the path connecting node 4 and node 1, and node 3 and node 2, and this value is the minimum unit of time distance based on the space distance to limit the search area. Min_Unit) (Step 102).

The starting node is referred to as node 5 (sp = 5), and the target node is referred to as node 9 (ep = 9) (Step 103).

The temporary shortest path and the shortest time distance at which the sum of the spatial distances from the starting node (node 5) to the target node (node 9) are minimum are calculated as follows (Step 104).

Nodes directly connected to the departure node (node 5) are nodes 2, 6, 7, 10, 11, and the distance from the departure node (node 5) to the arrival node (node 9) The node whose minimum spatial distance is minimum is node 7 (sum of spatial distance 5.16). The shortest distance (MINP_rdis) is 3.8, which is the time distance from node 5 to node 7 (rdis (5,7)), and the temporary shortest path is Tem_MINP [0] = 5 and Tem_MINP [1] = 7. .

Since node 7 is not the target node, node 7 searches again. Nodes 7 connected directly from node 7 and not included in the temporary shortest path are nodes 6, 9, and 10, and the dual departure node (5 The node where the sum of the space distance from node 1 and the target node (node 9) becomes the minimum is node 9 (sum of space distance 5.10). The temporary shortest path is Tem_MINP [0] = 5, Tem_MINP [1] = 7, Tem_MINP [2] = 9, and the shortest time distance MINP_rdis is 3.8 + 3.2 = 7.0.

The limit spatial distance (Lim_sdis) for reducing the search area is MINP_rdis / Min_Unit, which is 7.0 / 0.8929 = 7.84 (Step 105).

The node set NM within the critical clearance is composed of nodes whose sum of the spatial distance from the starting node sp of each node and the target node ep is smaller than the marginal clearance 7.84 (Step 106). .

NM = {5, 6, 7, 8, 9, 10, 11, 13}

For reference, the calculation results of the nodes included in the node set (NM) within the critical spatial distance from all nodes are shown in Table 3 below.

The minimum unit NMIN_Unit of the spatial distance for searching the shortest path in the set NM is 1.025, which is 1.025 (Step 107).

Search step ⅰ = 0, time distance from starting node c [0] = 0, search complete node set FN [0] = 5 (sp), temporary shortest path Tem_MINP [0] = 5 (sp), search node np [ 0] = 5 (sp) (Step 108).

Increase step 1 by one (Step 109)

Ⅰ = ⅰ + 1 (i = 1)

Node directly connected to node 5 which is np [0], not included in the temporary shortest path node (Tem_MINP [0] = 5), and not included in the search complete node set (FN [0] = {5}) Are nodes 6, 7, 10, and 11, and select node 6 (optional) (Step 110).

np [1] = 6

The selected node 6 is included in the search complete node set (FN [1] = {6}), and the time distance c [1] and the minimum expected time distance c_dis are calculated as follows (Step 112).

c [1] = c [0] + rdis (np [0], np [1]) = 0 + rdis (5, 6) = 0 + 4.1 = 4.1

c_dis = c [1] + sdis (np [1], ep) × NMIN_Unit = 4.1 + sdis (6,9) × 1.025 = 7.34

The minimum expected time distance (7.34) is greater than the shortest time distance (7.0), so go to Step 110 (Step 113).

A node directly connected to node 5 of np [0], not included in the temporary shortest path node (Tem_MINP [0] = 5), and not included in the search complete node set (FN [1] = {6}). Are nodes 7, 10 and 11, and select node 7 (Step 110).

np [1] = 7

The selected node 7 is included in the search complete node set (FN [1] = {6, 7}), and the time distance c [1] and the minimum expected time distance c_dis are calculated as follows (Step 112). ).

c [1] = c [0] + rdis (np [0], np [1]) = 0 + rdis (5, 7) = 0 + 3.8 = 3.8

c_dis = c [1] + sdis (np [1], ep) × NMIN_Unit = 3.8 + sdis (7,9) × 1.025 = 5.44

Since the minimum expected time distance (5.44) is smaller than the shortest time distance (7.0) (Step 113), the search node is stored in the temporary shortest path (Step 114).

Tem_MINP [1] = 7

Since search node 7 is not the target node 9, the flow proceeds to Step 109 (Step 115).

Increment step 1 by one (Step 109).

Ⅰ = ⅰ + 1 (i = 2)

Directly connected to node 7, which is np [1], not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 7), and the search complete node set (FN [2] = {} ) Nodes that are not included in) are nodes 6, 9, and 10, and select node 6 (Step 110).

np [2] = 6

The selected node 6 is included in the search complete node set (FN [2] = {6}), and the time distance c [2] and the minimum expected time distance c_dis are calculated as follows (Step 112).

c [2] = c [1] + rdis (np [1], np [2]) = 3.8 + rdis (7,6) = 3.8 + 1.5 = 5.3

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 5.3 + 3.16 * 1.025 = 8.54

The minimum expected distance (8.54) is greater than the shortest time distance (7.0), so go to Step 110 (Step 113).

Directly connected to node 7 which is np [1], not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 7), and completed searched nodeset (FN [2] = {6 }) Are not included in nodes 9 and 10, and select node 9 (Step 110).

np [2] = 9

The selected node 9 is included in the search complete node set (FN [2] = {6, 9}), and the time distance c [2] and the minimum expected time distance c_dis are calculated as follows (Step 112). ).

c [2] = c [1] + rdis (np [1], np [2]) = 3.8 + rdis (7,9) = 3.8 + 3.2 = 7.0

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 7.0 + 0 * 1.025 = 7.0

Since the minimum expected distance (7.0) is equal to the shortest time distance (7.0) (Step 113), the search node is stored in the temporary shortest path (Step 114).

Tem_MINP [2] = 9

Since search node 9 is the target node (Step 115), the shortest distance (MINP_rdis) becomes 7.0, which is the shortest time distance (c [2]), and the final step of the shortest path (Fi) is 2, Change the temporary shortest path to the shortest path from 0 to Fi (Step 116) and go to Step 110.

MINP [0] = Tem_MINP [0] = 5

MINP [1] = Tem_MINP [1] = 7

MINP [2] = Tem_MINP [2] = 9

It is directly connected to node 7 which is np [1] and is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 7) until the last step. ] = {6, 9}) is not included in node 10 (Step 110).

np [2] = 10

Include node 10 in the search complete node set (FN [2] = {6, 9, 10}), and calculate the time distance (c [2]) and the minimum expected time distance (c_dis) as follows (Step 112).

c [2] = c [1] + rdis (np [1], np [2]) = 3.8 + rdis (7,10) = 3.8 + 2.1 = 5.9

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 5.9 + 1.41 * 1.025 = 7.35

The minimum expected distance (7.35) is greater than the shortest distance (7.0), so go to Step 110 (Step 113).

Directly connected to node 7 which is np [1], not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 7) until the last step, and the searched node set (FN [2] ] = (6, 9, 10}), there are no nodes that are not included, so go to Step 117 (Steps 110 and 111).

Set the searched node set in the current step (ⅰ = 2) to an empty set (FN [2] = {}), and reduce the step to '1' (Step 117).

Ⅰ = ⅰ-1 (ⅰ = 1)

Step ⅰ is not '0', so go to Step 110 (Step 118).

Directly connected to node 5, np [0], not included in the temporary shortest path node (Tem_MINP [0] = 5) until the last step, and FN [1] = {6, 7} ) Are not included in nodes 10 and 11, and select node 10 (Step 110).

np [1] = 10

The selected node 10 is included in the search complete node set (FN [1] = {6, 7, 10}), and the time distance c [1] and the minimum expected time distance c_dis are calculated as follows. Step 112).

c [1] = c [0] + rdis (np [0], np [1]) = 0 + rdis (5, 10) = 0 + 4.1 = 4.1

c_dis = c [1] + sdis (np [1], ep) * NMIN_Unit = 4.1 + 1.41 * 1.025 = 5.55

Since the minimum expected time distance (5.55) is smaller than the shortest distance (7.0), the search node is stored in the temporary shortest path (Step 114).

Tem_MINP [1] = 10

Since the search node is not the target node, increase the step by '1' (Step 109).

Ⅰ = ⅰ + 1 (ⅰ = 2)

Directly connected to node 10, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and the search completed node set (FN [2] ] = {}) Is not included in nodes 7, nodes 9 and 13, and selects node 7 (Step 110).

np [2] = 7

The selected node 7 is included in the search complete node set (FN [2] = {7}), and the time distance c [2] and the minimum expected time distance c_dis are calculated as follows (Step 112).

c [2] = c [1] + rdis (np [1], np [2]) = 4.1 + rdis (10, 7) = 4.1 + 2.1 = 6.2

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 6.2 + 2.00 * 1.025 = 8.25

The minimum expected time distance (8.25) is greater than the shortest distance (7.0), so go to Step 110 (Step 113).

Directly connected to node 10, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and the search completed node set (FN [2] ] = {7}) are nodes 9 and 13, and node 9 is selected (Step 110).

np [2] = 9

The selected node 9 is included in the search complete node set (FN [2] = {7, 9}), and the time distance c [2] and the minimum expected time distance c_dis are calculated as follows (Step 112). ).

c [2] = c [1] + rdis (np [1], np [2]) = 4.1 + rdis (10, 9) = 4.1 + 2.8 = 6.9

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 6.9 + 0 * 1.025 = 6.9

Since the minimum expected distance (6.9) is smaller than the shortest time distance (7.0), the search node is stored in the temporary shortest path (Step 114).

Tem_MINP [2] = 9

Since search node 9 is the target node (Step 115), the shortest time distance (MINP_rdis) is changed to 6.9, which is the time distance (c [2]), and the last step (Fi) of the shortest path (Fi) is 2 at the current step,. Change the temporary shortest path to the shortest path from 0 to Fi (Step 116) and go to Step 110.

MINP [0] = Tem_MINP [0] = 5

MINP [1] = Tem_MINP [1] = 10

MINP [2] = Tem_MINP [2] = 9

Directly connected to node 10, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and the search completed node set (FN [2] ] = {7, 9}) is not included in node 13 (Step 110).

np [2] = 13

Include node 13 in the search complete node set (FN [2] = {7, 9, 13}), and calculate the time distance (c [2]) and the minimum expected time distance (c_dis) as follows (Step 112).

c [2] = c [1] + rdis (np [1], np [2]) = 4.1 + rdis (10, 13) = 4.1 + 2.0 = 6.1

c_dis = c [2] + sdis (np [2], ep) × NMIN_Unit = 6.1 + 2.24 * 1.025 = 8.40

The minimum expected time distance (8.40) is greater than the shortest distance (6.9), so go to Step 110 (Step 113).

Directly connected to node 10, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and the search completed node set (FN [2] ] = (7, 9, 13}), there are no nodes that are not included, so go to Step 117 (Steps 110 and 111).

Set the searched node set in the current step (ⅰ = 2) to an empty set (FN [2] = {}), and reduce the step to '1' (Step 117).

Ⅰ = ⅰ-1 (ⅰ = 1)

Step ⅰ is not zero, go to Step 110 (Step 118).

Directly connected to node 5, np [0], not included in the temporary shortest path node (Tem_MINP [0] = 5) until the last step, and FN [1] = {6, 7, 10}) is not included in node 11 (Step 110).

np [1] = 11

Include node 11 in the search complete node set (FN [1] = {6, 7, 10, 11}), and calculate the time distance (c [1]) and the minimum expected time distance (c_dis) as follows: (Step 112).

c [1] = c [0] + rdis (np [0], np [1]) = 0 + rdis (5, 11) = 0 + 2.8 = 2.8

c_dis = c [1] + sdis (np [1], ep) × NMIN_Unit = 2.8 + 3.61 * 1.025 = 6.50

Since the minimum expected time distance (6.50) is smaller than the shortest time distance (6.9), the search node is stored in the temporary shortest path (Step 114).

Tem_MINP [1] = 11

Since the search node is not the target node, increase the step by '1' (Step 109).

Ⅰ = ⅰ + 1 (ⅰ = 2)

Directly connected to node 11, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and the searched node set (FN [2] ] = {}) Is not included in node 13 (Step 110).

np [2] = 13

The node 13 is included in the search complete node set (FN [2] = {13}), and the time distance c [1] and the minimum expected time distance c_dis are calculated as follows (Step 112).

c [2] = c [1] + rdis (np [1], np [2]) = 2.8 + rdis (11, 13) = 2.8 + 3.6 = 6.4

c_dis = c [2] + sdis (np [1], ep) × NMIN_Unit = 6.4 + 2.24 * 1.025 = 8.70

The minimum expected time distance (8.70) is greater than the shortest time distance (6.9), so go to Step 110 (Step 113).

Directly connected to node 11, which is np [1], is not included in the temporary shortest path node (Tem_MINP [0] = 5, Tem_MINP [1] = 10) until the last step, and is a search completed node set (FN [2]. ] = {13}), so there are no nodes, so go to Step 117 (Steps 110 and 111).

Set the search completed node set in the current step (ⅰ = 2) to an empty set (FN [2] = {}), and reduce the step to '1' (Step 117).

Ⅰ = ⅰ-1 (ⅰ = 1)

Step ⅰ is not zero, go to Step 110 (Step 118).

Directly connected to node 5, np [0], not included in the temporary shortest path node (Tem_MINP [0] = 5) until the last step, and FN [1] = {6, 7, 10, 11}), there are no nodes that are not included, so go to Step 117 (Steps 110 and 111).

Set the search completed node set in the current step (ⅰ = 1) to an empty set ([FN1] = {}), and reduce the step to '1' (Step 117).

Ⅰ = ⅰ-1 (ⅰ = 0)

Since step ⅰ is 0 (Step 118), the shortest path (MINP [k], k = 0 to Fi) and the shortest time distance (MINP_rdis) are output.

Shortest path (MINP [0] = 5, MINP [1] = 10, MINP [2] = 9),

Shortest time distance (MINP_rdis = 6.9)

As described above, the present invention minimizes the search path of the shortest path and has the effect of accurately searching the shortest path to the target node in a short time. In addition, by using the unit of space according to the distance, it is possible to search not only the time distance but also the required (load) of any path such as actual distance and cost, and when the required time is required, It can also be applied to changes in volume, which enables the dynamic shortest path search.

Claims (5)

The command input unit such as a keyboard, mouse, and touch pad for selecting a start / destination node, a memory in which path information for a plurality of nodes is recorded, and a calculation unit using the memory information according to the input operation control signal In the method for limiting the search area for the path search, A first step in which the calculation unit is the minimum value of the actual required value divided by the spatial distance for all the links in the target network, based on the spatial distance; The operation unit establishes a temporary shortest path with nodes directly connected with the minimum of the spatial distance from the selected starting node and the target node through the command input unit, calculates the actual required value of the constructed temporary shortest path, and 2 steps to the minimum required value, A third step of calculating the marginal spatial distance obtained by dividing the shortest required value calculated in step 2 by the minimum circle unit based on the spatial distance obtained from the first step; 4. The method of claim 1, wherein the operation unit is configured to limit the search range by selecting nodes in which the sum of the spatial distances from the departure node and the arrival node is less than the limit spatial distance obtained in step 3. The method of claim 1, wherein the actual required value is selected from time, cost, and actual distance. A command input unit such as a keyboard, mouse, and touch pad for selecting a start / destination node, a memory in which path information of a plurality of nodes is recorded, and a calculation unit are minimized by using the memory information according to the input operation control signal. In the method of calculating the expected requirements, 1 step of calculating the minimum value of the actual distance required by the spatial distance based on the minimum value among the values obtained by dividing the actual required value by the spatial distance; For the specific node for which the calculation unit wants to know the minimum expected value, the actual distance value from the starting node to the specific node is calculated. A method for obtaining the minimum expected requirement value comprising the two steps of calculating the minimum expected value by summing the multiplied values. The calculation unit calculates the minimum value of the actual required values divided by the spatial distance for all the links of the network, and makes the value the minimum required unit based on the spatial distance, and the space from the starting node to the target node. 1 process of constructing the temporary shortest path with the nodes connected directly with the minimum distance sum and calculating the actual required value of the constructed temporary shortest path as the shortest required value; The operation unit sets the progress stage to 0, the actual distance to the progress stage is 0, includes the departure node in the searched node set in the progress stage 0, and sets the search node and the temporary shortest path node in the progress stage 0 as the departure node. With 2 courses, A third step of increasing the operation unit by one; 4 steps in which the operation unit is directly connected to the search node in the previous step and finds any node not included in the set of search completed nodes and the temporary shortest path node in the previous step, and makes a search node; If the operation unit does not have a search node in step 4, step 5 goes to step 9, If there is a search node in step 4, the operation unit stores the search completed node set in the current progress step and adds the actual required value from the previous search node to the current progress search node to the actual progress value in the previous progress step. The minimum expected value is calculated by updating the required value by adding the updated actual required value to the distance of the search node and the target node at the present stage and multiplying the minimum required unit based on the spatial distance obtained in step 1 6 courses, 7 steps in which the operation unit compares the calculated minimum expected value with the minimum required value and stores the search node in the temporary shortest path node if the minimum expected value calculated from the minimum required value is larger, and if it is smaller, If the minimum expected value is less than the minimum required value in step 7, the operation unit cycles to step 4 if the search node is not the target node and updates the actual required value updated to the shortest required value if it is the target node. 8 steps of converting the nodes stored in the temporary shortest path to the shortest path to the shortest path and then circulating to 4 steps, If there is no search node that satisfies the condition in step 4, the operation unit removes all nodes in the search completed node set in the current step and decreases the step by one. If the progress step is a non-zero progress step in step 9, the operation unit cycles to step 4, and if the progress step is 0, the shortest path search method characterized in that it consists of 10 processes to output the calculated shortest time distance and the shortest path nodes. The shortest path search method according to claim 4, wherein the shortest path search method uses the search path limiting method of claims 1 to 2, and obtains the minimum unit of space required based on the distance distance among the restricted nodes in step 1. Way.

Images