JPH06289929A - Optimum route deciding device - Google Patents

Abstract

PURPOSE:To easily decide a single optimum route even for regular courses. CONSTITUTION:A static graph generation part 10 calculates first costs for respective arcs connecting nodes which can be traveled based on the distances among the respective nodes in the course or shift time. When the target node is indicated, an angle potential calculation part 11 calculates second costs based on the directions of the respective nodes constituting the respective arcs viewed from the target node. The costs are added for the respective arcs. An optimum route generator 9 selects a case when the accumulated value of the added cost of the respective arcs becomes a minimum as the optimum route from the start node to the target node.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optimum route determining apparatus for determining an optimum route connecting a starting point and a target point in an unmanned transportation system such as a factory.

[0002]

2. Description of the Related Art As an optimum route determining device that requires less data setting in an unmanned transport system, the inventor of the present application has previously proposed a method using the configuration shown in FIG. 141373). FIG. 9 shows the configuration of the automatic guided vehicle. Reference numeral 16 denotes a map data memory in which map data of the traveling road is stored, and the coordinates of nodes at which the automatic guided vehicle can be stopped on the traveling road and their connection relationships are shown. Remembered Reference numeral 17 is an unmanned guided vehicle data memory in which data such as the speed of the unmanned guided vehicle is stored.

A graph generator 18 produces a graph G ₀ shown below. G ₀ = (N, A, C ₀ ) ... Here, N = {n ₁ , n ₂ , ..., N _m } is a set of nodes in which all nodes are numbered based on the map data, m
Is the number of nodes. A = {a ₁ , a ₂ , ..., A _n } has two adjacent nodes n _i , n _j as starting points and ending points, respectively.
When traveling between both nodes, it is a set of arcs in which all arcs a _k = {n _i , n _j } connecting both nodes are sequentially numbered, and n is the number of arcs. C ₀ is each arc a _k = {n _i , n _j } based on the cost calculation index such as the distance between the nodes.
Is a set of costs calculated for.

Numeral 19 is an optimum route generator which, when a transport instruction is sent from the control station (not shown) to this automatic guided vehicle, obtains a start node and a target node from the given transport instruction. Then, based on this and the graph G ₀ obtained by the graph generator 18, the map data and the unmanned guided vehicle data, an optimum route that minimizes the integrated cost is generated.

Here, as an index of cost calculation for each arc a _k = {n _i , n _j }, (a) the distance between nodes,
In addition to (b) travel time between nodes, (c) route directivity can be considered.

In the case of (a), the cost B _ij of the arc from the node i to the node j is obtained by the following equation (1). B _ij = d _ij (1) d _ij is the straight line distance (m between the start point node i and the end point node j
a m), is expressed as _{d ij = ((x j -x} i) 2 + (y j -y i) 2) 1/2 ...... (2). Here, x _i and y _i are x and y coordinates (mm) of the node i, and x _j and y _j are x and y coordinates (mm) of the node j.

In the case of (b), the cost B _ij of the arc from the node i to the node j is obtained by the following equation (3). B _ij = d _ij / v _ij (3) The distance d _ij is obtained in the same manner as above, and v _ij is the moving speed (m / sec) from the node i to the node j, so B _ij
Is an amount corresponding to the movement time between nodes.

In the case of (c), the cost B _ij of the arc from the node i to the node j is obtained by the following equation (4). B _ij = (d _ij / v _ij ) × (1-p _ij ) ... (4) The distance d _ij and the velocity v _ij are obtained in the same manner as above, and p _ij is “favorability” depending on the directionality of the route. Is a penalty coefficient. For example, if the direction from the node i to the node j is the "unfavorable" direction (reverse direction), p _{ij is set} to a negative number to increase the cost, while if it is the "favorable" direction (positive direction), p _ij is increased.
Can be made a positive number to reduce the cost. Coefficient p
The absolute value of _ij is set in the range of "0-1" according to the degree of preference. If appropriate penalty coefficients are set for all arcs, only one route (hereinafter referred to as the shortest route) that minimizes the integrated cost when connecting two arbitrary points (starting point and target point) is obtained.

[0009]

However, in order to properly set the penalty coefficient for all arcs, the influence of the penalty coefficient on the route search must be carefully considered, which is very complicated. . However, when the cost is calculated by the formula (1) or (3) that does not consider the penalty coefficient, the following problems occur.

In general, the traveling road is almost regular, such as the ladder type shown in FIG. 2 or the square lattice type shown in FIG. The ladder-type traveling path in FIG. 2 is composed of "28" nodes. Figure 6
(A) is (x, y) coordinate data of each node. FIG. 6B shows a summary of the node numbers, directions, and speed data of the start point and the end point for all the scenes that can move adjacently on this ladder-type traveling path. Here, the directionality of the route is not considered, and the direction data are all "0".
Has become.

Further, since the moving speed differs in the horizontal direction and the vertical direction, the cost is calculated by the equation (3) in consideration of the moving speed. FIG. 7A shows the calculation result beside each arc. Here, the costs of arcs whose start points and end points are opposite to each other are equal, and for example, the costs of arcs related to nodes 1 and 2 are B ₁₂ = d ₁₂ / v ₁₂ = ((4000 −1000) ² + (0 − 0) ² ) ^1/2
It is calculated as "3000" from / (1000/1000). Here, change the speed to "1.
The reason why it is divided by "000" is to match it with the unit of the equation (3). Similar calculations are performed for other arcs.

In order to determine the shortest route from the node 1 to the node 28 based on this data, it is sufficient to select a route that minimizes the total cost obtained by integrating the costs of the traveling arcs. However, here, in FIG.
As shown in (B), there are eight shortest routes with the same total cost.

The grid type traveling path of FIG. 3 is composed of "100" nodes. FIG. 8 shows (x,
y) Coordinate data. Here, the moving speeds between the nodes are all the same (1000 mm / sec), and the scene data is omitted. Furthermore, the distances between adjacent nodes are all 1
Since m is m, the cost of each arc calculated by the equation (1) or (3) is the same (1000).

Generally, in a lattice type running path of vertical p lines and horizontal q lines, the shortest number of routes from the lower left node to the upper right node is N (p, q) = _{p + q-2} C _p-1 . ... (5) As a result, when the shortest route from the node 1 to the node 100 in the traveling route of FIG. 3 is calculated,

[Equation 1] The route is selected as the same condition.

As described above, according to the calculation method based only on the distance between the nodes forming each arc or the traveling time, there are routes including an originally unfavorable direction, routes requiring frequent direction changes, etc. All are selected as the same condition. Therefore, it was difficult to determine a truly optimal route. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an optimum route determination device capable of easily determining an optimum route even on a regular traveling road.

[0016]

In order to solve the above problems, according to the present invention, a plurality of nodes scattered on a road are connected to adjacent first and second runnable nodes. With respect to a set of arcs, a first cost calculating means for calculating a first cost for each arc based on a distance or a moving time between the nodes, and a target node when the target node is instructed, The angle difference between the direction of the first node and the predetermined direction and the angle difference between the direction of the second node viewed from the target node and the predetermined direction are calculated, and for each arc based on the difference between the two angle differences. Second to calculate the second cost
Cost calculating means, the adding means for adding the cost calculating results by the first and second cost calculating means for each of the arcs, and the adding node calculated by the adding means based on the starting node to the target node. Path generation means for selecting a case where the integrated value of the added costs of the respective arcs is minimized as an optimum path to the above.

[0017]

In the above structure, the first cost calculating means calculates the first cost for each arc based on the distance between the nodes or the traveling time. When the target node is designated, the second cost calculation means causes an angular difference between the direction of the first node and the predetermined direction as viewed from the target node,
The angle difference between the direction of the second node viewed from the target node and the predetermined direction is calculated, and the second cost is calculated for each arc based on the difference between the two angles. Then, the adding means adds these costs for each arc. The route generation means selects a case where the integrated value of the added costs of the respective arcs becomes the minimum as the optimum route from the departure node to the target node.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an optimum route determination device according to the present embodiment, and the portions common to FIG. 9 are given the same reference numerals and the description thereof will be omitted.

In the figure, the graph generator 8 creates a graph G shown below. G = (N, A, C) ... where N = {n ₁ , n ₂ , ..., N _m } and A = {a ₁ , a ₂ ,.
a _n } is the same as each element of the graph G ₀ described above, and the description thereof will be omitted. In addition to the distance between nodes or the moving time described above, C is based on the index of cost calculation in which the angular potential described below is added, and each arc a _k = {n _i ,
It is a set of costs calculated for n _j }.

When the map data memory 16 is updated, the static graph generator 10 calculates the cost B _ij of each arc by the formula (1), (3) or (4).

When the target node g is designated, the angular potential calculator 11 calculates the angular potential cost T _ij (g) for each arc by the following equations (7) and (8).
To calculate. T _ij (g) = (K / 2π) Δθ (g, i, j) · B _ij (7) Δθ (g, i, j) = Mod (| θ (g, j) −θ (g, i) |, π) (8) Here, K is a predetermined coefficient, θ (g, i) is an angle obtained by measuring the direction of the node i viewed from the node g counterclockwise from the x axis, When g = i, it is set to "0". Also, equation (8) is
Shows the remainder when | θ (g, j) −θ (g, i) | is divided by π.

That is, the nodes i forming each arc,
For j, the direction of node i from the target node g and x
The angular difference from the axial direction and the angular difference between the direction of the node j viewed from the target node g and the x-axis direction are calculated. Then, the coefficient set based on the difference between the two angles is multiplied by the cost B _ij to calculate a new cost T _ij (g).

Then, the cost C _ij (g) shown below
(= C) is stored in the search graph data memory 12. C _ij (g) = B _ij + T _ij (g) (9) That is, the set of arc costs C _ij (g) is set depending on the target node in addition to the start and end nodes of each arc. It

The optimum route generator 9 receives the carrier instruction from the control station and determines the starting node and the target node as in the conventional case. Then, based on this and the above-mentioned graph G 1 obtained by the graph generator 8, map data, and unmanned guided vehicle data, an optimum route that minimizes the integrated cost is generated.

Consider the case where the target node is set to "28" in the ladder type traveling path of FIG. First, the static graph generation unit 10 causes the cost B _ij shown in FIG.
(Conventional cost) is created. Next, the angular potential calculation unit 11 calculates the angular potential cost T _ij (28) shown in FIG. Here, the coefficient K
Is "1". Then, for each arc, the cost B _ij and the angular potential cost T _ij (28) are added,
It is stored in the search graph data memory 12 as the cost C _ij (28). FIG. 4B shows the calculation result of the cost C _ij (28).

As a result, the shortest route from the node 1 to the node 28, which has been "8" in the conventional method, is "1 →
15 → 16 → 17 → 18 → 19 → 20 → 21 → 22 → 2
3 → 24 → 25 → 26 → 27 → 28 ”.

Consider also the case where the target node is set to "100" in the grid type traveling path of FIG. Also in this case, the cost B _ij and the angular potential cost T _ij (100) are calculated in the same manner as above, and the added cost C _ij (100) is stored in the search graph data memory 12. FIG. 5 shows the calculation result of the cost C _ij (100).

Therefore, in the conventional method, "4862"
The shortest path from node 1 to node 100 that existed as “0” is “1 → 11 → 21 → 31 → 41 → 51 → 61 → 7.
1 → 81 → 91 → 92 → 93 → 94 → 95 → 96 → 97
→ 98 → 99 → 100 ”is the only way to narrow down.

By the way, in the ladder type traveling path, the automatic guided vehicle moves mainly in the horizontal direction, and the work loading / unloading station is installed in the vertical direction of the traveling path. Further, when a large number of unmanned guided vehicles travel on such an elongated traveling path, it is less likely that they will interfere with each other if the unmanned guided vehicles travel as far outside as possible. Also, in the case of an automated guided vehicle that travels while measuring the distance to the vertical wall with ultrasonic waves,
Positioning at the station is assured by moving the distance in the horizontal direction as long as possible before the target node. In each of the shortest paths determined in the above-described embodiments, the horizontal or vertical arc tends to continue. In addition, it is easy to select a route along the outside of the travel route. This is a result of considering the angular potential cost, and therefore, it is possible to determine the shortest path that also satisfies the demand for the work environment as described above.

[0030]

As described above, according to the present invention, in addition to the first cost for each arc calculated based on the distance between nodes or the moving time, each arc seen from the target node Second based on the direction of each node that composes
Since the cost is calculated, it is possible to easily determine the optimum route even for a regular traveling road.

[Brief description of drawings]

FIG. 1 is a block diagram of an optimum route determination device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of a ladder type traveling path according to the present invention.

FIG. 3 is a diagram showing a configuration of a lattice type traveling road according to the present invention.

FIG. 4 is a diagram showing a result of cost calculation according to an embodiment of the present invention.

FIG. 5 is a diagram showing a result of cost calculation according to an embodiment of the present invention.

FIG. 6 is a diagram showing data such as coordinates of a ladder type traveling road according to the present invention.

FIG. 7 is a diagram showing conventional cost calculation and shortest path selection results.

FIG. 8 is a diagram showing coordinate data of a grid type traveling road according to the present invention.

FIG. 9 is a block diagram of a conventional optimum route determination device.

[Description of Reference Signs] 8 graph generator (adding means) 9 optimal path generator (path generating means) 10 static graph generating section (first cost calculating means) 11 angular potential calculating section (second cost calculating means)

Claims (1)

[Claims] 1. A set of arcs connecting adjacent first and second runnable nodes to a plurality of nodes scattered on a running road, and each arc is based on a distance or a travel time between the respective nodes. First cost calculation means for calculating the first cost for each, and when the target node is designated, the angular difference between the direction of the first node as viewed from the target node and the predetermined direction and the target node Second cost calculation means for calculating an angular difference between the direction of the second node and the predetermined direction, and calculating a second cost for each arc based on the difference between the both angles; Addition means for adding the cost calculation result by the second cost calculation means for each arc, and addition of each arc as an optimum route from the departure node to the target node based on the addition cost calculated by the addition means. Ko An optimal route determination device comprising: a route generation unit that selects a case where the integrated value of strikes is minimized.