JPH06273181A - Method and apparatus for deciding route, map managing apparatus and mobile body managing system - Google Patents

Abstract

PURPOSE:To provide a mean for deciding an optimum route responsive to a purpose based on a road map formed from an image map. CONSTITUTION:A distribution managing apparatus 1 has map managing means 7 and distribution plan drafting means 8. Image input means 9 reads a map, etc., sold in the market from an input unit 3, and stores it as image data (image map) in a memory 2. Road map forming means 10 displays the image map on a display unit 5, coordinate-converts a plurality of geographic points set on a screen (intersections, curved points of a road), and forms a road map formed of route data including positions, connections, directions. Route deciding means 12 reads the route data, first obtains a shortest route between two points of a starting point and a finishing point based on distribution information, traffic congestions, etc., from information input processing means 11, and further decides an optimum route (circulating route) of a plurality of distribution geographic points so as to satisfy designated target information (minimum distribution time/minimum designated time error difference, etc.).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for determining a route for a road network, and more particularly to a method for determining an optimal route at two points or an optimal route for efficiently traveling through multiple points, and a route search. The present invention relates to a map management device that generates useful map data and a mobile management system that efficiently manages mobiles.

[0002]

2. Description of the Related Art In conventional navigation and delivery plans,
A database that is a digital (also called vector) of a commercially available map (image) is used. According to this, since the road network is divided at appropriate intervals and the coordinates of the plurality of connection points are provided, it is possible to calculate the distance and search for the shortest route.

However, in determining a route such as a delivery plan, generally, a specific visiting point group in a specific area is followed by a route that is determined to some extent in the shortest time, at the minimum energy (cost), or at a designated time. It is necessary to make the deviation to the minimum of.

Therefore, in formulating a delivery plan, a deformed map (may be handwritten) within the area required by the user is much more useful. Further, while a detailed map contains much unnecessary data, it is necessary to update the information frequently because it is detailed, which causes unnecessary work and cost.

Therefore, as in the map data creating apparatus described in Japanese Patent Laid-Open No. 63-10275 (Reference Example 1), the coordinates of a plurality of points on the road are converted into coordinates, and the length of a straight line between the two points is calculated. There is one that edits the so-called vectorized data, which is converted into the value of the orthogonal coordinate system normalized by.

On the other hand, in order to deal with the explosion of the combination of routes, the operation principle of the neural network is applied like the neural network type shortest route search device of Japanese Patent Laid-Open No. 4-247555 (Reference Example 2). There is also a method to find the shortest route.

[0007]

However, in the reference example 1, the road map along the road on the image map can be created, but the traffic control information and the traffic jam information cannot be provided.
It is insufficient as data for route determination.

Since the reference example 2 is based on the Hopfield type neural network, the solution is at an extreme value and the optimum solution cannot be obtained, as is well known. Also,
Since the objective function is a second- to third-order mathematical expression, an extremely large number of calculations are required, and the processability is extremely low.

In the delivery plan based on the same model, a certain degree of solution can be obtained in a limited time at 10 to 20 points even for a simple map, and 10 points or less for a complicated map.

Further, it is considered that the purpose of the plan is only the shortest time or the shortest distance, and it is not possible to deal with the minimum specified time deviation.

An object of the present invention is to overcome the above problems of the prior art and achieve the following.

First, to provide a route determination method and apparatus for generating route data suitable for route search from an image map and using this to easily determine a route having a shorter distance or less time. It is in.

Secondly, there is provided a route determining method and device for rapidly determining an optimal route according to various purposes such as a delivery plan based on route data including the traveling direction of the road and dynamic conditions. To do.

Thirdly, to provide a map management device for inputting an arbitrary image map including a handwritten map, simply creating route data suitable for the purpose of use, and creating a road map. is there.

A fourth object is to provide a mobile object management system capable of efficiently operating the mobile objects and improving the service by searching the routes of a plurality of mobile objects corresponding to the request.

Other objects of the present invention will be clarified by the following description.

[0017]

In order to achieve the above-mentioned object, the present invention has a constitution comprising the following features, respectively. The first feature of the present invention is that a plurality of points are connected to each other. In a route determining method or device for determining a predetermined route in a road network that exists, specify a plurality of points on the road network of the input image map and establish a connection relationship with the position data of those points. Route data including distance data between certain points is created, two or more predetermined points are designated from the points based on the created route data, and a plurality of points are connected to each other. The purpose is to search the route of and determine the route satisfying the objective information of the shortest distance or the shortest time.

A second feature of the present invention is, in a route determining method or apparatus in a road network, position data of a plurality of points in the road network and distance data between points having a connection relationship with each other.
Based on the route data including the data, two or more predetermined points are designated from the above points, a plurality of routes connecting these predetermined points are searched, and the optimum route satisfying the predetermined target information is determined. To do.

The route search is performed by creating a predetermined objective function for a vector obtained by randomly arranging the plurality of predetermined points to obtain a first objective function value, and then arranging the nodes of the vectors. Is calculated based on two kinds of uniform random numbers to obtain a second objective function value, and it is determined which of the first objective function value and the second objective function value is closer to the optimal solution. Then, the candidate route in the vector sequence closer to the optimal solution is set as the optimal route candidate, and this process is repeated.

A third feature of the present invention is a map management device, which is an image input means for inputting an image map and converting it into image data, and an image map displayed on the screen. For a plurality of points arbitrarily designated above, route data including position data of the points, distance data between the points having a connection relationship with each other, and direction data for a predetermined point.
Means for generating a road map, a means for storing the image data of the image map, a storage means for storing the route data, and a display for displaying the image map and / or the road map. And means.

A fourth feature of the present invention is a mobile object management system, which is a route data including position data of a plurality of points of a road network and distance data between points having a connection relationship with each other. The shortest distance between the road map management means that manages the data, the position detection means that acquires the current position of the moving body, the requested position that requests the service by the moving body, and the positions of the plurality of moving bodies. A moving body to be a route or a shortest time route is determined based on the route data, and a moving body determination instruction means for giving an instruction to the moving body and a position and a required position of the moving body on the road map are displayed. And output means for outputting.

[0022]

According to the first feature of the present invention, a commercially available image map is read, and a plurality of points on the road network designated on the screen are connected to the position data of those points and are connected to each other. The shortest distance or the minimum time of a predetermined point specified based on the route data is created by creating distance data between certain points and further direction data (traffic control information). You can search and decide the route of.

As a result, the route search can be performed easily and at a low cost without using the conventional enormous and expensive database.

According to a second aspect of the invention, for example:
When applied to a delivery planning device, first, the optimum route between all delivery points on the designated map is determined. This is to replace the route from one delivery point (start point) to another delivery point (end point) with a route tree with the start point as a root (root), and when the end point is connected to the tree, the route is the optimum route. The above process is repeatedly performed while perturbing the order of the examination point nodes, a result close to the optimum solution is stochastically determined, and the optimum route (for example, the shortest route) between the two points is determined.

Next, the order and route of the designated delivery points are determined so that the designated objective function value is optimum. This is to calculate the objective function when visiting the specified delivery point based on the optimal route between the two points determined between each delivery point and the set road / traffic information. , The delivery point examination order is iteratively perturbed and the optimal solution is determined stochastically.

According to this, the explosion of the combination problem caused by a large number of paths consisting of a large number of points can be eliminated, and the optimum solution of the maximum value (minimum value) instead of the extreme value can be determined in a short time, and It is possible to determine the optimum route corresponding to various conventionally difficult purposes such as the minimum specified time deviation of the delivery point.

According to the third feature of the present invention, arbitrary map data such as handwriting is input and the image displayed on the screen is displayed.
Route map for multiple points arbitrarily specified on the map
It is possible to display or print out a desired road map. Furthermore, by providing the shortest route data between all two points in the predetermined area obtained in advance, it is possible to quickly and easily perform the navigation for outputting the optimum route only by instructing the start point and the end point.

According to the fourth aspect of the present invention, in dispatching a vehicle such as a taxi, the shortest distance or the shortest time can be reached by searching for a route between a required point and a plurality of serviceable taxis. Vehicle dispatch instructions can be automatically determined, and efficient operation of multiple mobile units and improvement of service can be achieved.

[0029]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a first embodiment according to the present invention and shows a configuration of a delivery plan management apparatus which is an application example of a route determination apparatus.

The delivery plan management device 1 reads an image map from the image input device 3, creates digital (vector) road map data, stores it in the storage device 3, and a setting device. The delivery plan planning means 8 determines the optimum delivery route based on the delivery information from the vehicle 4 and the road map data and outputs it to the display device 5 and the printing device 6.

FIG. 2 schematically shows a hard structure for realizing the delivery plan management system. The management device 1 includes a central processing unit (CPU), RO (not shown).
It can be realized by a computer including M, RAM, input / output I / O, and a bus connecting these. The storage device 2 is the RAM
And an auxiliary storage device, and an optical or magnetic disk device or the like can be used as the auxiliary storage device.

An image scanner is used as the image input device 3, but it may be a reading device (for example, a disk drive) of a storage device which stores image image information.

As the setting device 4, for example, a locator or keyboard such as a mouse, or a touch panel attached to the display screen is used. Also, a setting device with a receiving device, which takes in remote sensing information (for example, mobile body position information by GPS) and the like, can be used.

Next, the structure and operation of the map management means 7 will be described.

The map management means 7 comprises an image input means 9 and a road map creation means 10 as shown in FIG. The image input means 9 inputs an arbitrary map from the image scanner 3 and stores it in the image map storage area 51 of the storage device 2 as image data. This image data is represented by a 1/0 signal in a predetermined pixel unit, and these are realized by a known technique.

The road map creating means 10 reads the image map from the storage device 3, displays it on the display device 5, and converts it into digital data (vector data) according to the route information instructed on the screen. The generated road map data is generated.

FIG. 3 shows an example of an image map in which a commercially available map is read, temporarily stored in the storage means 5, and again read out and displayed on the screen. A predetermined area of a certain city is reduced to 1/20000.
It is represented by.

FIG. 4 is a flow chart showing the processing procedure of the road map creating means 10. The operation will be described with reference to FIGS.

The user picks an intersection, a bending point or a specific point on the road in FIG. 3 with a mouse on the screen. Each designated point is converted into coordinates on the screen, and the designated number and coordinates are stored as the point information table 52 shown in FIG. 5 (step A).

Next, the route is designated and the distance is calculated by the designation (step B). Designation of the route is performed in three ways (a) to (c) as shown in FIG.

First, the most common bidirectional routing is performed. This is specified in [1] ~ as shown in FIG.
Two points you want to route from the point [4], for example,
Pick [1] and [2]. As a result, point [1]
The bidirectional designation between [1] and [2] is fetched, and then the distance between them is calculated.

The distance calculation is performed by the flow shown in FIG. In step B10, the coordinates of the points [1] and [2] are fetched from the point information table 52, and the distance is calculated in the screen coordinates. Next, the scale of the image map is input or read and converted into an actual distance (step B20).

After that, the route definition table 53 shown in FIG.
Store in the row → column and column → row areas of. In the example of the figure, 13 are provided in both areas of [1] → [2] and [2] → [1].
km is stored. The above steps B10 to B30 are performed for all points to be routed.

Next, in step C of FIG. 4, one-way traffic is designated. For example, between points [2] and [3],
Points [2], [3] where bidirectional routes have already been specified according to the route information of "one way from [2] to [3]"
Are designated in the order of [2] → [3] on the screen. As a result, the distance data in the [3] → [2] area of the route definition table 53 is erased.

Next, in step D of FIG. 4, a right / left turn prohibited route is designated. As shown in FIG. 6C, right / left turn prohibition is possible by designating three points. Figure 9 shows
In the figure explaining the left turn prohibition designation, at point [2],
When prohibiting the right turn of [1] → [4], the order is [1] → [2] → [4] on the screen. This allows
The prohibition flag "1" is set in the (1, 2, 4) area of the right / left turn prohibition information table 54. Similarly,
The left turn prohibition of [1] → [2] → [7] is (1, 2, 7)
"1" is set in the area.

For the above-mentioned route direction designation, that is, one-way traffic, right / left turn prohibition, etc., if this information is stored in advance by linking it to a point, the step is linked with the process of step B of FIG. C and D can be automatically performed.

The right / left turn prohibition information table 54 is
The point (j in FIG. 9) which is compressed only to the address having the prohibition flag and serves as the turning point is defined in the route definition table.
Attach it to the corresponding point on Bull 53. According to this, in the route search to be described later, it is possible to refer to the right / left turn prohibition information by reading the route table.

As described above, the map managing means 7 generates the route information table 53 and the road map data mainly including the right / left turn prohibition information table. FIG. 10 shows an example in which this road map data is displayed in an overlapping manner on the image map. In this example, the point is 23 of [1] to [239].
Nine points are designated ([1] in the specification is indicated as, but they point to the same point), and usable routes are designated as indicated by thick black lines. Needless to say, one-way traffic, right / left turn prohibition, and the like can also be displayed in a distinguishable manner by the line type, color, and the like.

Further, although the point on the road on the image map screen is designated in the above, a point not on the road on the screen can be designated. In this case, roads not shown on the image map, planned roads, etc. can be arbitrarily added.

In the route definition table 53, the actual distance is stored in the area of the matrix in which the point number is defined on both axes. Writing to and reading from this table are performed by the variable PATH (m, n). Here, m and n are the corresponding two points.

[0052]

## EQU1 ## PATH (1,7) = 27 (1)

[0053]

[Formula 2] PATH (7,1) = ъ (2) Equation (1) shows that there is a route from the point [1] to the point [7], and the distance is 27 (km). . On the other hand, in the expression (2), there is no route from the point [7] to the point [1], and there is one-way from the point [1] to the point [7].

The route definition table 53 is stored in the computer as a network definition table 55 shown in FIG. 11 in order to speed up the processing. Node number on the vertical axis,
The number of paths linked to the corresponding node and the link destination node are defined on the horizontal axis, and are referred to by NETDEF (i, j).

Next, the structure and operation of the delivery plan making means 8 in FIG. 1 will be described.

The delivery information / external information input processing means 11 is
Delivery information such as delivery position, delivery volume, delivery time, and purpose information of delivery plan, dynamic conditions of road network (for example, traffic jam condition, sudden traffic blockage, etc.), time of delivery vehicle position information, etc. The external information that changes is input, processed such as predetermined calculation and editing, and stored in the storage device 2.

The route determination means 12 determines a delivery route based on the planning instruction information, the road map data and the delivery information, and outputs it to the display device 5 and the printing device 6.

FIG. 12 shows an example of delivery information. In this example, the starting point is [11] at 8:00,
It is necessary to deliver the set amount of load to each of the 16 points shown in the figure and return to the point [12]. Delivery times are points [62], [141], [114], [105], [1]
99], but no other points are specified. This delivery information is stored in the storage device 2 as the delivery information table 55 by the delivery information input means 11.

FIG. 13 shows the traffic volume (congestion level) of the two routes.
The time pattern of is shown.

The pattern is that normal driving is possible from 0:00 to 6:00, and the degree of traffic is 1.0
Has become. From 7:00 to 10:00, traffic jams occur during the commuting hours and the traffic volume has risen to 2.8. A similar traffic jam occurred from 16:00 to 20:00 in the evening.

In the pattern, the traffic volume is high and the traffic volume level is high during the daytime from 11:00 to 17:00, but is hardly affected by commuting.

A plurality of patterns as described above are formed for each route and day of the week, and are made into tables by the external information input means 11 so as to be selectable between respective points and registered in the storage means 5. This is used to calculate the travel time of the predetermined route.

For example, if the pattern [1] is set for the route from point [1] to [7] in FIG. 10, an attempt is made to pass through this route at 9:00 when the traffic volume is 2.8. In this case, it takes 2.8 times as long as when passing at 5:00 when the traffic volume is 1.0.

FIG. 14 shows an example of objective function information. In this example, as the optimization items in the plan, “minimum delivery time”, “energy-minimum”, “minimum designated time deviation”,
One can be selected (*) from five types of "minimum delivery distance" and "maximum delivery time margin". The objective function information is stored in the storage device 2 as the objective function information table 56 by the delivery information input processing means 11.

As the objective function information, "minimum delivery time" is selected when there are many delivery points and it is expected that the ratio of delivery time to standard operating time will be large. On the other hand, if you have a lot of time, but the amount of delivery is large and you want to minimize the fuel consumption of the delivery means (vehicle), "Minimum energy"
Is selected.

The energy E in this case is, for example,

[0067]

[Equation 3] E = Σ Wi · Li (i = 1, ..., N) (3) Here, Wi is the load weight from point [i] to point [i + 1], and Li is the distance from point [i] to point i + 1.

With this definition, a plan is created in which the delivery points with a large delivery amount are preferentially selected,
Total energy is minimized. This energy calculation is performed by the delivery information input processing means 11. The details of each objective function will be described later.

The delivery management device 1 includes road map data by the map management means 7 and delivery information / external information input processing means 1.
The delivery information and external information according to No. 1 are provided as described above.

Therefore, it is activated at an arbitrary timing by the instruction information from the setting device 4, and the route determining means 12 makes an optimum delivery plan as follows.

FIG. 15 is a flowchart showing the processing procedure for determining the optimum delivery route by the route determining means 7. Here, the shortest route between delivery points is determined (a) to (c).
Then, it is roughly divided into steps (d) to (d) for determining the delivery route.

The search for the shortest route is the optimum route when the objective function is the minimum distance route. However, the shortest route between points is essential basic data even in route determination by other objective functions. Therefore, if the shortest route between all target points is made into data, the optimum delivery route according to the purpose can be determined at high speed.

<1> Determination of Shortest Route between Delivery Points First, steps (a) to (c) will be described. In FIG.
Steps (a) and (b) are processing steps for repeatedly setting all routes starting from all points and ending at all points. Start point pointer I for iterative processing
Then, 1 to n delivery points are set in the end point pointer J, respectively. Step (c) is the optimal route between points I and J,
Here, the shortest route is determined.

The processing procedure for determining the shortest route will be roughly described with reference to the flowchart for determining the optimum route shown in FIG. 16, and the details thereof will be described later.

First, in processing step A, a vector X in which all points are randomly arranged is generated. Next, in step B, a logical tree having the starting node as a root is generated in accordance with the node arrangement order in the vector X. At this time, when the end point node appears on the leaf, the route is saved as the route length. Let this be f (X).

Next, when the number of nodes is N, N ² times, D ~
The process of G is repeated.

In step D, a part of the forward vector X of the node is slightly changed to generate a new vector Y.
In step E, the logical tree for the vector Y is generated by the same procedure as in step B, and the path length f (Y) is calculated.

In step F, the path lengths of the combination vectors X and Y are compared with each other, and the one closest to the optimum solution is set as a candidate for the optimum solution (step G), and the process returns to step D.

The judgment of the above step F is invented by the following facts.

In FIG. 17, the horizontal axis defines the forward vector of the node, and the vertical axis defines the sum of the distances between the two points. In the case of the vector A, the total distance is 465, and the total distance of the vector OPT which is a little away is 380, and the vector OPT gives the optimum solution.

Assuming that the total distance of the vector B is 480, which is a slight modification of the vector A, the vector A is obviously smaller and excellent for the purpose of the minimum distance. It is possible that they will be removed from the candidates. However, when A is at the minimum point, it is impossible to find the one that exceeds A by a slight change in the vector, so that the optimum solution cannot be reached forever.

In the present invention, in order to get out of the minimum value in such a case, the concept of the temperature T, which is reduced according to the number of examinations, was introduced to solve the problem. The temperature T here does not mean a physical temperature.

That is, when the minimum force required for the vector A to reach the OPT after passing the vector B without fail is T1, and the minimum force for getting over the mountain of the vector B from the OPT is T2,

[0084]

## EQU00004 ## T is set so that T1.ltoreq.T <T2 (5). For example, if the number of searches is i
And

[0085]

[Equation 5] T = Δ / log (i + c) (6) (Δ is a positive real constant, log is a natural logarithm, and c is a positive real constant). It gets smaller as it gets bigger.

From the above, the determination condition of step F is, for example,

[0087]

## EQU6 ## where a <exp (-(f (Y) -f (X)) / T) (7) (a is a uniformly distributed random number, exp is the exponentiation of natural logarithm), and Equation (7) ) Is satisfied, even if f
Even if (Y) is inferior to f (X), the forward vector Y may be adopted as a combination close to the optimal solution.

Next, the above-mentioned optimum route determination method will be described in detail by taking a simplified network as shown in FIG. 18 as an example. In this example, the starting point node I is the point.

[0], the end point node J is the point [16], and [1] to [1
5] and are joined together. The number between each point is the actual distance between the points. This value is read from the route definition table 53 or from the route cost table 58 shown in FIG. 19, which is a table of the cost between points according to the objective function information. It is assumed that there is a bidirectional route between each point and there is no right / left turn prohibition.

For convenience of description, first, the processing flow of the route search / evaluation (steps B and E in FIG. 16) shown in FIG. 20 will be described. This is to generate a logical tree (tree) from the starting point node while checking the shortest distance to reach the ending point node.

First, the randomly arranged route vectors are
X2 = ([1], [2], [7], [3], [16],
[14], [15], [10],

[9], [8], [1
1], [12], [13], [4], [5], [6])
, And is given.

The vector X2 is the combination vector X (p) of the tree-work table shown in FIGS.
It is stored in.

In step a, the tree work table TRWK (l) of FIG. 28 is initialized (for example, "-1"),
It is set from the examination node P = 1.

Next, in step b, MIN ← ∞ (path length work variable), PNODE ← -1 (optimum path parent node)
Set.

In step c, the number of paths connected to the node P is set to the network definition table NE in FIG.
Obtained from TDEF (i, j), the following examination is performed.

Since P = 1 now, the node linked to node [1] is

Starting point from [0], [4], [5]

The shortest route from [0] is determined in steps d and e, and the node [1] is set as shown in (1) of FIG.

It is connected to [0] (steps f and g).

Since the node [1] is not the end point, the process P ← P + 1 is performed in step i, and the process returns to step b.

At this time, the state in the work area is as shown in FIG.
As shown in 7, the parent node of node 1 is

At [0], the path length from the starting point is 115.

Similar processing is performed by P = 4, that is, node 3
Repeated until, (4) of FIG. 23 was obtained at this point,
The contents of the work area are as shown in FIG.

When P = 5 and the node becomes [16], none of the paths linked to [16] are connected to the tree from the origin. Therefore, [16] is connected after the last element of the route vector, and the value of P is updated.

This also applies to the following nodes [14] and [15], and the work state of the route vector at the time when the examination of the node [15] is completed is X2 =
(1, 2, 7, 3, 16, 14, 15, 10, 9, 8,
11,12,13,4,5,6,16,14,15).

This is X2 = (1,2,7,3,10,9,8,1) as a logical tree forming this network.
1,12,13,4,5,6,16,14,15). In other words, only the vector that holds as a logical tree is searched without waste.

When the above operation is repeated until P = 17, (P
Figure 2 shows the route selection in the middle of the process up to = 17.
4, FIG. 25, FIG. 26), and FIG. 27, the end point node [16] is connected to the logical tree

The route from [0] is established.

The work table at this time is as shown in FIG.

The route length from [0] to node [16] is 415, and the route is

[0] → [2]
→ [7] → [10] → [16].

It is clear that the above process gives the shortest path connecting the starting point and the ending point in the logical tree created in the order of the path vector X2. Also, one node is at most "C
Assuming that it has 1 ”nodes, C1 × (N + C2)
It is possible to generate the shortest route by studying (times).
Here, C2 is the number of nodes that could not be connected to the tree in the examination process and are stored in the stack that holds the arrangement order.

Generally, since C2 <C3 × N, the total number of examinations is C1 × (N + C2) <C1 × (N + C
3 × N) = C1 × ((1 + C3) × N) = C4 × N (however, C1, C2, C3, C4 << N).

This shows that for a sufficiently large N, the shortest route of the starting point and the ending point in the logical tree generated by a certain route vector can be determined by the number of examinations in the order of N.

In the above processing, if one-way traffic, right / left turn prohibition, etc. are considered, the following processing is performed.

That is, each time a new vector sequence is generated, the route definition table 53 and the right / left turn prohibition information table 54 are referred to. The vector search is stopped, a new vector is generated again, and the search is continued.

Next, regarding the processing flow of FIG. 16, FIG.
1 and FIG. 22. In step A,
As a random initial vector, X1 = (1,2,8,
9,10,15,14,16,3,7,11,12,1
3, 4, 5, 6) are arranged.

Then, in step B, the above-described processing of FIG. 20 is performed, and the optimum route in X1 is

[0] →
[2] → [8] →

It is generated as [9] → [14] → [16].

From FIG. 19, the route cost C ¹ (P
_(1)-(16) ) is

[0112]

[Equation 7] C ¹ (P _(1)-(16) ) = C (P _(1)-(2) ) + C (P _(2)-(8) ) + C (P _(8)-(9) ) + C (P _(9)-(14) ) + C (P _(14)-(15) ) = 145 + 68 + 60 + 73 + 39 = 385 (8) However, the cost in this example is the shortest actual distance of each route.

Next, the processing from steps D to G is performed N times.
Repeated ^twice.

At step D, the contents of the route vector X1 are slightly changed as follows.

First, in step D, the number of nodes is 1 to 16
Random numbers i, j (i <
j) is generated.

Next, i in the order of arrangement of the elements of the vector of X1
A new route vector Y (= X2) is created by reversing the order from the th to the jth.

In this example, since i = 3 and j = 10 (FIG. 21), the arrangement of the vector elements from the third node [8] to the tenth node [7] is changed to Y (= X2) = (1,2,7,3,16,1
4,15,10,9,8,11,12,13,4,5,
6) has been obtained (for the sake of explanation, it is the same as the X2 vector searched above).

Next, in step E, the above-mentioned processing is performed,
The optimum route for X2 shown in FIG.

[0] →
[2] → [7] → [10] → [16] is determined. The cost in this case is “415” as described above, and X
It is larger than the cost for 1.

In step F, X1 and X2 (= Y) are compared as follows, and a process for determining a route vector close to the optimum solution is performed.

First, a real random number a uniformly distributed between 0.0 and 1.0 is generated. Here, for example, a =
It is assumed that 0.750 is obtained.

Next, assuming that Δ = 100.0 and c = 2 in the relational expression for determining the temperature, that is, the above expression (6), i
= 1 (i is the number of searches), T = 100 / lo
g (1 + 2) = 209.5.

Next, from the formula (7), exp (-(41
5-385) /209.5) = 0.867> a = 0.7
50.

As a result, the cost of X2 (= Y) is X1.
It is adopted as a vector of route candidates that gives the optimum solution, although it is larger than

In the next iterative process, FIG.
As shown in

[0] → [2] → [7] → [1
The route 0] → [16] is determined as the optimal route for X3, and the same result as X2 is obtained.

Since this satisfies the judgment condition in step F, X3 becomes a new optimum solution candidate this time.

By repeating the above processing, the route to be obtained gradually approaches the optimum solution. In this example, as shown in FIG.
As shown in, the optimal solution is reached after only five searches.

As shown in FIG. 22, once the optimum solution is reached, no other state is entered. That is, X6 to X7 in FIGS. 22 to 22 are not judged to be close to the optimum solution with respect to X5.

As described above, according to this embodiment, the optimum route (in this example, the shortest route) can be determined by approaching the optimum solution to be obtained with a small number of searches without stopping at the extreme value. .

In FIG. 31, the shortest route search between two points ([62] → [105]) is performed by determining the optimum route, and the shortest route is deformed on the image map. It is displayed. Suitable for navigation etc.,
It is self-explanatory.

<2> Formulation of Optimal Delivery Plan Here again, the process flow of the delivery plan formulation shown in FIG.
Returning to, the procedure for formulating the optimal delivery plan will be explained.

Next, in steps (d) to (d), the optimum delivery route that optimizes the objective function value is determined.

FIG. 32 shows the target road network of the delivery plan. In this example, the point [11] is departed, the following 16 points are visited, and the points [12] are returned to [31], [62]. ],
[56], [141], [151], [114], [9
1], [8], [83], [105], [177],
[166], [199], [220], [225],
The delivery route of [237] is assumed.

Therefore, the route is 16! = 16x
There are 15 × …… ... × 1> 20 × 10 ¹² types, and even if a high-speed computer is used, the enumeration method for examining all of them takes a long time and is not practical. Moreover, in reality, there are several tens to several hundreds of delivery points, so conventionally, there was no choice but to rely on an empirical engineering method, and the optimal solution decision has been abandoned.

First, in step (d), the visit order is randomly created, X = (31, 62, 56, 141, 15).
1,114,91,8,83,105,177,16
6,199,220,225,237) to the vector X
And

Although there are a plurality of points and routes which are not visited between the adjacent delivery points in X, as described above, in steps (a) to (c), the shortest distance between the two points (start point to end point) is set. A route is obtained, and as shown in FIG. 33, a shortest route table 59 having each delivery point as an end point is created for each origin node ([11] in this example). By selecting the start point and end point, the shortest route and actual distance between each delivery point can be immediately indexed.

The thick black circles and solid lines in FIG. 32 represent the vector X.
It shows the actual route on the network created based on. This is the optimum delivery route in the arrangement of the vectors X. Strictly speaking, it is an optimized route in a sequence of Xs that minimizes the cost of the target information (minimum time in this example) based on the shortest route, which will be described in detail later.

At the step (e), the planning examination repeating process pointer K is set to 1 to the cube of the number of delivery points, that is, 16 ³ = 4096.

In step (f), since the plan vector Y is generated by slightly changing the vector X, 1 to 1
Two kinds of random numbers i1, i2 (i1 which are integers in the range of 6)
<I2) is generated. Here, when i1 = 2 and i2 = 6, all the visit orders of the second to sixth visits of the vector X are exchanged. This is an operation equivalent to the mutation operator of gene direction reversal in the genetic algorithm.

As a result, Y = (31, 114, 151,
141, 56, 62, 91, 8, 83, 105, 17
7,166,199,220,225,237) are generated.

[0140] Y has changed five of the elements constituting X. Considering the delivery route, the point [11
4] and [62] are only replaced, and they are processed so as not to make a big change. This is i1, i
This is because, no matter what value 2 is, the number of visiting points is changed to two.

At step (g), the objective function values of the newly generated plan vectors Y and X, F (Y) and F (Y), are obtained.
(X) is calculated. In this example, the objective function is "minimum delivery time",

[0142]

## EQU8 ## F (Y) = Σ Ti (i = 1, ..., N) (9)

Here, Ti is the time required to move from the point [i] to [i + 1], and is calculated from the moving distance per unit time of the transportation means and the traffic volume at each time of each route (FIG. 13). Desired.

The definition of the objective function information shown in FIG. 14 will be described below.

In the case of "minimum energy",

[0146]

(9) F (Y) = Σ Wm · Ti (i = 1, ..., N) (10) Here, Wm is a load when moving from the point [i] to [i + 1].

In case of "minimum specified time deviation",

[0148]

F (Y) = Σ (THi-TPi) ² (i = 1, ..., N) (11) Here, THi is the delivery time at point [i], TP
i is the planned arrival time of [i]. This formula is a quantification of the time shift with respect to the desired delivery time.

"Minimum delivery distance" is equivalent to "minimum delivery time" when traffic conditions such as traffic congestion are not considered,

[0150]

(11) F (Y) = Σ Li (i = 1, ..., N) (12) Here, Li is the actual distance from the point [i] to [i + 1] and can be obtained by indexing the shortest route table in FIG.

In the case of "maximum delivery time allowance", there is a need to deliver the designated delivery time as early as possible.

[0152]

F (Y) = Σ (THi-TPi) (i = 1, ..., N) (13) may be maximized. THi and TPi are the same as the “minimum designated time deviation”.

As described above, by defining the objective function corresponding to various needs, the optimum solution of the delivery plan desired by the user can be easily determined.

In step (h), the fully open optimal plan candidate vector X and the vector Y obtained by slightly changing X,
Determine which is closer to the optimal solution.

FIG. 34 conceptually shows the method of determining the optimum value (OPT), taking the transition of the objective function values of “delivery time”, “energy”, and “designated time deviation” as an example.

Taking the delivery time as an example, the objective function value of the vector X is F (X) = 320, and the objective function value of the vector Y is F (Y) = 340. Is superior in delivery time, and X is more likely to be closer to the optimal solution.

However, if such a judgment is made, the optimum solution is not reached. This is an experimentally clarified fact. As shown in the enlarged portion of the figure, once the extreme value X is generated in the process of obtaining the optimum solution vector OPT, the vector is changed slightly. This is because the new vector cannot be the optimal value candidate because it cannot escape from the extreme value.

Therefore, the next maximum point C may be overcome from the objective function value F (X) of the vector X giving the minimum value. Therefore, the temperature T that becomes smaller as the number of examinations increases is defined, and the above-described mathematical expressions (5) to (7) are used.
To change the judgment standard.

For example, in this example, F (X) = 32
0, F (Y) = 340, Δ = 4.0, T = 4.0
/Log(1+2)≈3.6, a = 0.00400, exp (− (F (Y) −F (X)) / T = exp
(-(340-320) /3.64) = exp (-(2
Since 0 / 3.64) ≈0.00411> a, Expression (7) holds.

Therefore, the vector Y is replaced with X ← Y in step (i) as a candidate on the route leading to the optimum solution, and becomes a new optimum solution candidate.

As described above, the optimum solution can be reliably reached by repeating steps (d) to (ri) in FIG. 15 three times the number of delivery points.

The number of repetitions is a value obtained by many experiments, and no matter what state the examination is started, it is sufficient to examine the cube of the number of planned objects at most.

FIG. 35 shows the result of the above-mentioned processing. The minimum time delivery route, OPT = ([3
1], [62], [56], [237], [225],
[220], [199], [177], [166],
[151], [141], [114], [105],
[83], [91], [8]) are shown.

The delivery route shown in FIG. 35 is designed so that the delivery time is the shortest in consideration of the traffic conditions and the road conditions which change with time. That is, as compared with the route of FIG. 33, for example, the most congested point [141],
It is possible to visit all delivery points without waste while making the time of visiting [151] and the like relatively long in the daytime.

[0165] When "Minimum energy" is selected as the objective function, for example, if the points with a large amount of delivery are [237] and [91], these are given priority and other delivery points on the route. Efficient delivery routes are planned, which will be visited while stopping by.

FIG. 44 shows an example in which the optimum delivery route is displayed by being superimposed on the image map, and the target information (in this example, "shortest route within the delivery date") is also displayed. If this screen is converted into a paper map by the printing device 6, it can be used as it is for the operation of the delivery vehicle.

The determination of the optimum route described above can be realized by using neuro or simulated annealing or a genetic algorithm, but other operating methods can be used.
It is also possible to use a techniques research or an artificial intelligence method.

Next, a second embodiment of the present invention will be described.

When the delivery plan determined by the delivery management apparatus 1 of FIG. 1 is being executed and the plan is out of order due to an unexpected accident or traffic jam, it is necessary to review it according to the actual situation.

FIG. 36 shows a processing flow in the case of modifying the delivery plan being executed.

[0171] Now, every time the vehicle travels along the optimum delivery route and is delivered at the delivery point, the delivery means reports the arrival point and time to the delivery management device 1. This report may be made by telephone, or automatically detected by the GPS (artificial satellite position management system) or the like.

When these are input to the external information input processing means 11 via the setting device 4 (step A), the delivery information table 56 (FIG. 12) is searched for the delivery time scheduled time at the delivery point. It is determined whether the deviation is a certain value or more (step B).

As a result, if the value is equal to or more than a certain value, re-planning is requested for the remaining delivery points starting from the delivery point to satisfy the delivery information (step C). The replanning is started automatically by the delivery plan information input means 11. Of course, the user may start.

Next, a map management system according to the third embodiment of the present invention will be described.

FIG. 37 shows the structure of the map management system. The map management device 21 has the same basic structure as the map management means 7 of FIG.

FIG. 38 shows a processing flow by the map management device 21. A process for creating a road map shown in the lower part of FIG. 39 from an image map having two different scales shown in the upper part of FIG. 39 will be described according to this flow.

At step A, each point including the factory is designated, and the coordinates of the designated point are set. In step B, the distance between the two points is calculated on the screen, and in step C, a representative scale (scale) of the map, for example, 1/100.
It is converted to an actual distance by 00 and set in the route definition table in step F. The processing up to this point is the same as the processing in FIG. 7.

Next, in step D, if other scales are set, in step E, the two points designated in the area are designated and converted to the actual distances of other scales. , Set to the route definition table. Furthermore, steps G and H
Then, set one-way and left / right turn prohibition.

By the way, in this example, the urban area is 1/10000.
The area near the factory is 1/5000
This is the case when the scale is 0. Moreover, in this area, some of the roads required for route determination are newly built on the input image map, so it is not always shown on the image map. It includes cases where it does not exist. That is,
The designated point is not always on the road network of the image map.

However, according to the map management apparatus of the present embodiment, even in such a case, a point arbitrarily set on the road or outside the road and the route data corresponding to the scale of the area are generated and desired. You can create a road map of.

Therefore, it is suitable for the case of navigating a desired optimum route on the basis of a handwritten map in which a part of a region is deformed and drawn.

Further, the map management / management device 21 of this embodiment.
Can be provided with a plurality of shortest route tables 59 (FIG. 33) at two points, which are calculated in advance for a predetermined area.

According to this, the user can obtain the shortest route navigation as shown in FIG. 31 only by designating the start point and the end point.

Next, a mobile unit management system according to a fourth embodiment of the present invention will be described.

FIG. 40 shows a mobile unit management system, and the mobile unit management apparatus 31 is constructed as shown in the block diagram of FIG.

The position information management device 32 shown in FIGS. 40 and 41.
Displays and manages the current position of the vehicle and the vacant state,
It is also possible to make a vehicle dispatch request and a vehicle pick-up instruction. The mobile unit instruction determination means 35 determines the most appropriate mobile unit selection instruction for the request based on the information and the request from the position information management device 32.

FIG. 42 shows the processing flow of the mobile unit instruction determining means 35 by taking the taxi dispatch service as an example. In Step A, the position of the currently operating vehicle and the vacant vehicle status are known.

FIG. 43 shows this situation and shows the service of a predetermined area visualized by the display device 5 or the position management device 22.
This is the operational status of taxis that can be screwed. At the point of ~, there is an empty car in the direction of the arrow.

At step B, the passenger waiting state, that is,
Check whether or not there is a dispatch request. Now, it is assumed that a user's vehicle allocation request occurs at the double circled point in FIG. In step C, the shortest distance between the required point and the vehicle is calculated and stored as a vehicle allocation candidate (step E). The shortest distance of each vehicle is sequentially calculated, the vehicle closest to the required point is selected (step D), and the vehicle dispatch instruction is output to the vehicle closest to the required point (step F).

The vehicle to be dispatched may be limited to within a predetermined radius from the request point. In addition, each point on the road map that serves as the basis for distance calculation is provided for each lane to consider the traveling direction, or is managed by a line based on the point, that is, a zone (zone) in the length direction of the road. You may do it.

Further, if the shortest route data between two points described in the above embodiment is stored in the storage device 2, higher speed processing becomes possible.

[0192]

According to the route determining method and device of the present invention,
Since the route data required for the search can be created from the image map and the route having a smaller distance or time can be determined, the route search can be performed easily and inexpensively using a commercially available map or the like, which is expensive.

According to the route determination method and apparatus of the present invention, the optimal route can be determined by the route data suitable for the search and the fast search of the optimal solution using the route data. The shortest route between two points can be obtained in an extremely short time, facilitating navigation and the like.

(2) Optimal delivery (visit) routes at a large number of delivery points can be obtained according to the purpose such as the shortest time / designated time, and the delivery plan can be easily planned.

(3) Since the external information such as the traffic situation and the progress of the plan is referred to for the route determination, it is possible to flexibly make a plan and make a correction corresponding to the changing external situation.

According to the map management apparatus of the present invention, a road map including route data including connection between points, distance, and direction can be created by specifying image map data and route designation. 1) A user can easily create a desired road map convenient for route determination from a commercially available map or a handwritten map.

(2) Since the shortest route information between a plurality of two points is attached to the route data, the shortest route can be immediately output and displayed only by designating the desired two points.

According to the mobile unit management system of the present invention, since the map intermediary device having the route data and the position management unit for the mobile unit are provided, the mobile unit on the shortest route is selected from the request point. Therefore, there is an effect of efficiently managing a plurality of moving bodies and improving service.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a delivery management device according to a first embodiment of the present invention.

FIG. 2 is a hard block diagram of a delivery plan planning device.

FIG. 3 is a screen display diagram of an image map.

FIG. 4 is a flowchart showing a processing procedure for generating route data.
To.

[Figure 5] Point information table that stores the coordinate data of the designated point
building.

FIG. 6 is an explanatory diagram for explaining a method of adding bidirectionality, unidirectionality, and right / left turn prohibition directionality to route data.

FIG. 7 is a flowchart showing a procedure for capturing an actual distance between points.

FIG. 8 is a table showing an example of a route definition table PATH.
Bull.

FIG. 9 is a conceptual diagram illustrating right / left turn prohibition information table and its setting.

FIG. 10 is a screen display diagram in which a road map based on route data is superimposed on an image map.

FIG. 11 is a network definition table converted from the route definition table.

FIG. 12 is a table for storing delivery information.

FIG. 13 is a traffic volume / time graph showing patterns of traffic conditions by route.

FIG. 14 is a table for storing objective function information.

FIG. 15 is a flowchart showing a procedure for planning a delivery management device.

FIG. 16 is a flowchart showing a processing procedure of an optimum (shortest) route determination method between two points.

FIG. 17 is an explanatory diagram for explaining a convergence situation to an optimum route.

FIG. 18 is a simulated network diagram for explaining an optimum route determination method between two points.

FIG. 19 is a cost table obtained by converting a route definition table by objective function information.

FIG. 20 is a flowchart showing a processing procedure of route search / evaluation logic.

FIG. 21 is a schematic diagram conceptually explaining the examination transition of the optimum route determination.

FIG. 22 is a schematic diagram conceptually explaining the examination transition of the optimum route determination.

FIG. 23 is a schematic diagram conceptually explaining a study transition of route search / evaluation.

FIG. 24 is a schematic diagram conceptually explaining a study transition of route search / evaluation.

FIG. 25 is a schematic diagram conceptually explaining a route search / evaluation study transition.

FIG. 26 is a schematic diagram conceptually explaining a study transition of route search / evaluation.

FIG. 27 is a schematic diagram conceptually explaining a study transition of route search / evaluation.

FIG. 28 is a diagram showing a tree generation situation of route search / evaluation.
Table.

FIG. 29 is a diagram showing a tree generation status of route search / evaluation.
Table.

FIG. 30 is a diagram showing a tree generation status of route search / evaluation.
Table.

FIG. 31 is an example of a screen in which the shortest route search result between two points is identified and displayed on an image map.

FIG. 32 is a display example of a route based on a vector initialized by the delivery planning means.

FIG. 33 is an example of the shortest route table (node [11]) generated by the optimum route determination logic.

FIG. 34 is an explanatory diagram illustrating the transition of each objective function value of delivery time, energy, and designated time deviation, and the principle of optimal solution convergence.

FIG. 35 is a screen display diagram of the optimum route based on the vector finally determined by the delivery plan making means.

FIG. 36 is a flowchart for processing re-planning during plan execution according to the second embodiment of the present invention.

FIG. 37 is a block diagram showing the configuration of a map management device according to a second embodiment of the present invention.

FIG. 38 is a flowchart for explaining the operation of the map management device.
To.

FIG. 39 is a screen display diagram showing a handwritten map and a road map created by the map management apparatus.

FIG. 40 is a block diagram showing the configuration of a mobile management system according to a third embodiment of the present invention.

FIG. 41 is a block diagram showing a concrete station configuration of the mobile management device.

FIG. 42 is a flowchart showing the processing procedure of the mobile unit management device.

FIG. 43 is a screen display diagram illustrating Taxi vehicle allocation, which is an application example of a mobile unit management device.

FIG. 44 is a screen display diagram in which the optimum delivery route is superimposed and displayed on an image map.

[Explanation of symbols]

1 ... Delivery management device, 2 ... Storage device, 3 ... Image input device, 4 ... Setting device, 5 ... Display device, 6 ... Printing device, 7 ...
Map management means, 8 ... Delivery planning means, 9 ... Image input means, 10 ... Road map creation means, 11 ... Delivery information / external information input processing means, 12 ... Route determination means, 21 ... Map management device, 31 ... mobile object management device, 32 ... position management device, 33 ... mobile object management means, 34 ... mobile object management information input processing means, 35 ... mobile object instruction determining means.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hideo Yoshida 3-2-1, Saiwaicho, Hitachi-shi, Ibaraki Hitachi Engineering Co., Ltd. (72) Inventor Yasuhiro Terada 5-2-1, Omika-cho, Hitachi-shi, Ibaraki No. 1 Incorporated company Hitachi Ltd. Omika Plant (72) Inventor Yoshiyuki Sato 52-1 Omika-cho, Hitachi City, Ibaraki Prefecture Incorporated Hitachi Ltd. Omika Plant (72) Inventor Masakazu Yahiro Omika-cho, Hitachi City, Ibaraki Prefecture 5-2-1, Hitachi Ltd. Omika Plant (72) Inventor Akemi Otsuki 4-6 Kanda Surugadai, Chiyoda-ku, Tokyo Hitachi Ltd.

Claims (27)

[Claims] 1. A method of determining a predetermined route in a road network in which a plurality of points are connected to each other, wherein a plurality of points are designated on the road network of an input image map, and the points are designated. And position data, and route data including distance data between points that are in a connection relationship with each other are created, and based on the created route data, two or more predetermined points are selected from the points. Is specified, a plurality of routes connecting these predetermined points are searched, and a route satisfying the objective information of the shortest distance or the minimum time is determined. 2. A method for determining a predetermined route in a road network in which a plurality of points are connected to each other. In the method, the position data of a plurality of points in the road network and the distance between the points having a connection relationship with each other. Based on the route data including the data,
A route determination method comprising: designating two or more predetermined points from the points, searching for a plurality of routes connecting the predetermined points, and determining an optimum route satisfying predetermined target information. 3. The route determining method according to claim 1, wherein the route data has direction data such as one-way traffic between points and prohibition of right / left turn. 4. A route determination method according to claim 1, 2 or 3, wherein said route data also includes dynamic information which changes with time. 5. The method according to claim 1, 2 or 3 or 4, wherein the predetermined points are two points which are a start point and an end point or a plurality of visiting points to be visited, and the determined route or the optimum route is the A route determination method, which is a route from a start point to an end point (two-point route) or a route that passes through all of the plurality of visited points (tour route). 6. The target information according to claim 2, wherein the determined route has a minimum distance, a minimum travel time, or a minimum movement energy as compared with other routes.
Minimum or specified time deviation Minimum or maximum delivery time,
Which is to set any of the, the optimal route, corresponding to the set target information,
A method for determining a route, wherein a value of an objective function that takes a sum of cost between points of a route under consideration is a minimum or maximum. 7. The route determining method according to claim 6, wherein when the target information is the minimum distance, the cost between points is the shortest distance data between the predetermined points. 8. The route determination method according to claim 6, wherein when the target information is a minimum route travel time, the point-to-point cost is a minimum travel time between the predetermined points. 9. The minimum movement time according to claim 8,
A route determination method characterized in that the route information is corrected according to the degree of traffic that changes over time. 10. The method according to any one of claims 6 to 8,
For the route search, for a vector in which the plurality of predetermined points are randomly arranged, a predetermined objective function is created to obtain a first objective function value, and then two kinds of the node arrangement of the vector are obtained. The second objective function value is obtained for the new vector changed based on the uniform random number of, and it is determined which of the first objective function value and the second objective function value is closer to the optimal solution, The candidate route in the vector sequence closer to the solution is set as the optimal route candidate,
A route determination method characterized by repeating this process. 11. The method for determining a path close to the optimum solution according to claim 10, wherein the temperature decreases as the number of iterations increases, and a temperature T that determines the energy of a random system is set. And the value fol of the first objective function
d, the values fnew and T of the second objective function are α <e
xp (-(fnew-fold) / T) (exp represents the power of the base of natural logarithm)
The candidate route of the new vector is defined as an optimal route candidate,
A route determination method characterized in that iterative processing is performed a predetermined number of times. 12. The method according to claim 11, wherein the number of times of the iterative processing is C.n ³ times (C), where n is the number of target nodes (points).
Is a constant) and can be set within the range. 13. A road network in which a plurality of points are connected to each other, and in a device for determining a predetermined optimum route according to target information, there is a connection relationship with position data of a plurality of points in the road network. Map management means for creating and managing route data having distance data between points, a function for reading the route data in a predetermined road network (network), and all the nodes thereof. A function of determining a candidate route by arranging (points) at random, creating a logical tree having a predetermined starting point as a route (root), and connecting a predetermined end point to the logical tree. A function for changing the node sequence of the candidate route based on two kinds of uniform random numbers, and a logical tree for rooting the start point for the changed node sequence is created and the end point is created. A new route sequence connected to the logical tree as a new candidate route, and The objective function defined on the basis of the objective information, the candidate route and the new candidate route is calculated, a function for obtaining both objective function values, and which objective function value is closer to the optimal solution is determined, A route determining means having a function of using a route route candidate route closer to the optimal solution as an optimal route candidate, and a function of repeating these processes of defining the candidate routes and determining a route closer to the optimal solution; A route determination device comprising: 14. The route determining means according to claim 13, wherein the route determining means has a function of performing an iterative process so as to generate a logical tree in the order in which the nodes are arranged, and the node to be examined is a starting point node. If a node can be connected to a root-based logical tree, it is connected to the logical tree.
If the node is the end point, the route is determined as the optimum route and the process is terminated. If the node is not the end point, the process proceeds to the next process, and the node under consideration routes the start point node. If it cannot be connected to the logical tree that is the (root), the relevant node is the last node.
A routing device having a function of storing as a code and proceeding to the next processing. 15. The route determining device according to claim 14, wherein the route determining means determines each shortest route between a plurality of two points each including a combination of a predetermined start point and a predetermined end point. . 16. The route determination device according to claim 14 or 15, wherein setting means for designating predetermined two points from among the plurality of two points, and the shortest route between the two points designated by the setting means. A route determination device comprising output means for displaying and / or printing. 17. An apparatus for determining a route according to predetermined purpose information in a road network in which a plurality of points are connected to each other, wherein position data of a plurality of points in the road network are connected to each other. Road map management means for managing route data having distance data between certain points, and two or more predetermined points are designated from the points based on the route data, and these predetermined points are designated. It comprises a planning means for searching for a plurality of routes connecting the points and determining an optimum route satisfying a predetermined purpose information, wherein the planning means is a predetermined road network (network), For all two points consisting of a combination of a start point and a predetermined end point, a function for determining and storing the shortest route, and a vector in which a plurality of visiting points are randomly arranged in the predetermined road network (network). The function to generate and Between each visiting point of the vector, the function of creating a predetermined objective function by making a predetermined objective function based on the shortest route and the arrangement of the vector are changed based on two kinds of uniform random numbers. A function of obtaining a second objective function value for a new vector, determining which of the first objective function value and the second objective function value is closer to the optimal solution, and the vector sequence closer to the optimal solution. Function to make the candidate route as the optimum route candidate, a function to repeat the process of generating this vector and the determination of the optimum route candidate, and a function to determine the optimum visit route according to the target information and make a predetermined visit plan. And a function of outputting the planned plan including the optimum visiting route, the route determining device. 18. The route determining device according to claim 17, further comprising means for acquiring the execution status of the visit plan, and when the deviation of the execution status acquired at a predetermined visiting point exceeds a predetermined value, the path planning apparatus starts from the point. And a planning unit having a function of instructing re-planning of the route. 19. An image input means for inputting an image map and converting it into image data, and a plurality of points arbitrarily designated on the image map displayed on the screen, said points. Road map creating means for generating route data including position data of the above, distance data between points having a connection relationship with each other, and direction data about a predetermined point, said image A map management device comprising: the image data of a map; a storage unit for storing the route data; and a display unit for displaying the image map and / or the road map. . 20. The map according to claim 19, wherein the road map creating means calculates distance data between the points corresponding to a plurality of scales of the input image map. Management device. 21. The map management device according to claim 19 or 20, wherein the direction data indicates one-way traffic, right turn, and left turn prohibition. 22. The map management device according to claim 21, wherein the display means displays the direction data of the road map in a distinguishable manner. 23. The map management device according to claim 19, further comprising route determining means for determining an optimal route between a predetermined start point and end point based on the route data. 2. A map management device characterized in that the optimum route obtained between the two points designated by is displayed on the display means so as to be superimposed on the image map. 24. The route data according to claim 23, wherein the route data additionally has dynamic information such as a traffic volume of the route which fluctuates with time, and the route determining means has the two designated points. A map management device characterized by obtaining a shortest time route between them. 25. The map management according to any one of claims 19 to 24, wherein the arbitrarily designated point is on and / or outside the road network of the image map. apparatus. 26. Route data including position data of a plurality of points on a road network and distance data between points having a connection relationship with each other.
The minimum distance between the road map management means that manages the vehicle, the position detection means that obtains the current position of the moving body, the required position where the moving body is requesting the service, and the positions of multiple moving bodies. A moving body to be a route or a shortest time route is determined based on the route data,
A mobile body management system comprising: a mobile body determination instruction means for giving an instruction to the mobile body; and an output means for outputting a position of the mobile body and a requested position on the road map. 27. The mobile management system according to claim 25, wherein the mobile is a taxi.