JP7458593B2 - Method, system and program for creating travel routes for each of a plurality of moving objects - Google Patents

Description

The present invention relates to a method, system, and program for creating movement routes for each of a plurality of moving objects.

2. Description of the Related Art Autonomous flying vehicles (eg, drones) are used in a wide variety of applications, such as logistics, inspection, civil engineering/architecture, and aerial photography.

Patent document 1 discloses a system that dynamically plans paths for autonomous mobile robots, including drones.

Special Publication No. 2020-502630

The inventor of the present invention thought that by simultaneously using a plurality of moving objects such as autonomously navigable flying objects or autonomously movable robots, the range of applications of these moving objects could be further expanded.

However, there is no established method for automatically creating safe paths that minimize collisions between multiple moving objects.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a method, system, and program for creating movement routes for a plurality of moving objects to minimize collisions between the plurality of moving objects. .

The present invention provides, for example, the following items.
(Item 1)
A method for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the method comprising:
(1) A first starting point and a first plurality of candidate arrival points for the first moving body, and a second starting point and a second plurality of candidate arrival points for the second moving body. obtaining movement conditions including;
(2) determining a straight path for each of the plurality of moving objects,
determining a first linear route of the first moving body by connecting the first starting point and one first selected candidate destination point of the first plurality of candidate destination points; and,
determining a second linear route of the second moving body by connecting the second starting point and one second selected candidate arrival point of the second plurality of candidate arrival points; determining a straight path, including;
(3) for at least the first straight route, the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point; deriving a first modified path by connecting .
(Item 2)
(2) Determining the straight line path includes determining each of the first straight line path and the second straight line path so that the intersection between the first straight line path and the second straight line path is minimized. The preparation method described in item 1, comprising determining.
(Item 3)
The creation method according to item 2, wherein the intersection is an intersection on a two-dimensional plane.
(Item 4)
(2) Determining the straight path is as follows:
identifying a straight line that does not intersect with any other straight line path;
A point group including the first starting point, the second starting point, and a candidate arrival point group including the first plurality of candidate arrival points and the second plurality of candidate arrival points is dividing by a straight line,
Identifying the first selection candidate destination point within the point group that includes the first starting point among the point group after division;
The creation according to any one of items 1 to 3, including specifying the second selection candidate destination point within the point group that includes the second starting point among the point group after division. Method.
(Item 5)
The creation method according to any one of items 1 to 4, wherein the at least one point is a point in a point group arranged around the first straight line path.
(Item 6)
The plurality of points in the point group are distributed such that the closer the points are to the first starting point, the denser the points are, and/or the closer the points are to the first selected candidate destination point, the denser the points are. The preparation method described in item 5.
(Item 7)
The point group is
Obtaining a point cloud with a random distribution in a polar coordinate system;
Converting a point group in the polar coordinate system to a point group in a rectangular coordinate system;
centering the transformed point cloud on the first starting point;
The creation method according to item 5 or item 6, wherein the converted point group is arranged by arranging the point group centered on the first selection candidate arrival point.
(Item 8)
Deriving the first modified path includes:
(i) calculating a repulsive force field exerted by an obstacle on the first straight path;
(ii) connecting the first starting point, the first selected candidate destination, and the at least one point so that the repulsive force field is minimized. The preparation method described in paragraph (1).
(Item 9)
Deriving the first modified path includes:
(a) calculating a repulsive force field exerted by an obstacle on the second straight path;
(b) based on the repulsive force field, at least one of the second starting point, the second selected candidate destination point, and the second selected candidate destination point between the second starting point and the second selected candidate destination point; deriving a second modified path by connecting the points;
(c) deriving the first modified route by performing (i) and (ii) on the first straight route using the second modified route as an obstacle; , the preparation method described in item 8.
(Item 10)
The creation method according to any one of items 1 to 9, wherein the first moving object and the second moving object are randomly selected from the plurality of moving objects.
(Item 11)
A system for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the system comprising:
Movement including a first starting point and a first plurality of candidate destination points for the first mobile object, and a second starting point and a second plurality of candidate destination points for the second mobile object. an acquisition means for acquiring conditions;
Determining means for determining a straight path for each of the plurality of moving objects,
determining a first linear route of the first moving body by connecting the first starting point and one first selected candidate destination point of the first plurality of candidate destination points; and,
determining a second linear route of the second moving body by connecting the second starting point and one second selected candidate arrival point of the second plurality of candidate arrival points; a determining means configured to:
connecting the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point for at least a first straight path; and deriving means for deriving a first modified path by.
(Item 12)
A program for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the program being executed in a computer system having a processor section, The program is
(1) A first starting point and a first plurality of candidate arrival points for the first moving body, and a second starting point and a second plurality of candidate arrival points for the second moving body. obtaining movement conditions including;
(2) determining a straight path for each of the plurality of moving objects,
determining a first linear route of the first moving body by connecting the first starting point and one first selected candidate destination point of the first plurality of candidate destination points; and,
determining a second straight path of the second moving body by connecting the second starting point and one second selected candidate arrival point of the second plurality of candidate arrival points; determining a straight path, including;
(3) for at least the first straight route, the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point; A program that causes the processor unit to perform processing including: deriving a first modified path by connecting .

According to the present invention, it is possible to provide a method, system, and program for creating movement paths for multiple moving bodies that minimizes collisions between the multiple moving bodies. By following these movement paths, the multiple moving bodies can move safely with collisions minimized.

FIG. 1 is a diagram showing an example of a flow for using multiple flying objects 10. A diagram illustrating an example of the configuration of a system 100 for creating movement routes for each of a plurality of moving objects. A diagram showing an example of the configuration of the processor section 120 Diagram explaining depth-first search method with pruning that counts the number of pair intersections FIG. 1 is a schematic diagram showing the steps of the Sequential Partition method. Diagram showing schematically how to arrange a point cloud around a straight path Diagram showing schematically how to arrange a point cloud around a straight path FIG. 1 is a schematic diagram showing a flow of identifying at least one point between a starting point and a selected candidate destination point by the A* algorithm and deriving a modified path. A diagram schematically showing the flow of identifying at least one point between the starting point and the selected candidate destination point using the A* algorithm and deriving a modified route. A diagram schematically showing the flow of identifying at least one point between the starting point and the selected candidate destination point using the A* algorithm and deriving a modified route. A diagram schematically showing the flow of identifying at least one point between the starting point and the selected candidate destination point using the A* algorithm and deriving a modified route. A diagram schematically showing the flow of identifying at least one point between the starting point and the selected candidate destination point using the A* algorithm and deriving a modified route. Flowchart illustrating an example of a process 700 by the system 100 for creating movement routes for each of a plurality of moving objects. Flowchart showing an example of a process 800 for arranging a point group around a straight path connecting a starting point and a selected candidate destination point 9 is a flow chart illustrating an example of a process 900 for deriving a modified path. A flowchart showing an example of a flow for planning the movement routes of each of a plurality of moving objects using the system 100 for creating the movement routes of each of a plurality of moving objects. An example of a user interface 1100 of an application for implementing the process 700 for creating movement routes for each of a plurality of moving objects Flowchart illustrating an example of an application operating method for implementing processing 700 for creating movement routes for each of a plurality of moving objects. An example of a screen 1300 including a 3D display of a flight path presented to a user

Embodiments of the present invention will be described below with reference to the drawings.

As used herein, "about" refers to a range of ±10% of the following number.

1. Flow using multiple flying objects FIG. 1 schematically shows an example of a flow using multiple flying objects 10 as an embodiment of the mobile object.

Here, a plurality of flying objects 10 (for example, unmanned aerial vehicles) are used at destination G (for example, to observe the weather at the destination, measure the environment at the destination, and deliver cargo to the destination. An example will be explained in which a plurality of flying objects 10 are autonomously navigated from a departure point S to a destination G for transportation, etc.).

In step S1, the user U inputs navigation conditions into the computer system 100 as initial settings. The navigation conditions include at least information on the departure point S and destination G. The information on the departure point S is expressed, for example, as position coordinates of the departure point of each of the plurality of flying objects 10. The information on the destination G is expressed, for example, as the position coordinates of the arrival point of each of the plurality of flying objects 10. The navigation conditions may also include information such as the number of aircraft to be navigated and the navigation range.

When the navigation conditions are input, the computer system 100 creates a navigation route for each of the plurality of flying objects 10 based on the navigation conditions. At this time, the computer system 100 can create a navigation route in which the respective navigation routes of the plurality of flying objects 10 minimally interfere with each other.

In step S2, the computer system 100 instructs each of the multiple flying objects 10 to navigate according to the created navigation route.

In step S3, each of the multiple flying objects 10 navigates from the departure point S to the destination G according to the respective navigation routes created by the computer system 100. Because the multiple navigation routes created by the computer system 100 have minimal interference, each of the multiple flying objects 10 can navigate safely from the departure point S to the destination G.

When the multiple aircraft 10 reach destination G, in step S4, the multiple aircraft 10 can be used for tasks according to their intended use. For example, the weather at destination G is observed using observation equipment mounted on the multiple aircraft 10. For example, the environment at destination G is measured using measuring equipment mounted on the multiple aircraft 10. For example, luggage loaded on the multiple aircraft 10 is unloaded at destination G.

In this way, the navigation route created by the computer system 100 enables safe navigation from the departure point S to the destination G, and promotes the use of a plurality of flying objects 10 at the destination G. Thereby, the range of uses of the plurality of aircraft 10 at the destination G can be expanded.

In the above example, an air vehicle 10 has been described as a moving body, but the present invention is not limited to air vehicles and can be applied to any moving body. Any moving body includes, but is not limited to, a moving body that moves on the ground, a moving body that moves underground, a moving body that moves on water, a moving body that moves underwater, a moving body that moves in outer space, and the like. In addition, the moving body may be capable of moving autonomously without a control input, or may be capable of moving by a control input. The control input may be input, for example, by an operator on board the moving body, by an operator outside the moving body, or by a computer system outside the moving body.

The computer system 100 described above can be implemented by a system for creating travel routes for each of a plurality of moving objects, which will be described below.

2. Configuration of a system for creating travel routes for each of a plurality of moving bodies FIG. 2 shows an example of the configuration of a system 100 for creating travel routes for each of a plurality of moving bodies.

The system 100 includes an interface section 110, a processor section 120, and a memory section. The system 100 is connected to a database section 200.

The interface unit 110 exchanges information with the outside of the system 100. The processor unit 120 of the system 100 can receive information from outside the system 100 via the interface unit 110, and can send information to the outside of the system 100. The interface unit 110 can exchange information in any format.

The interface unit 110 includes, for example, an input unit that allows information to be input into the system 100. It does not matter in what manner the input unit allows information to be input into the system 100. For example, if the input unit is a touch panel, the user may input information by touching the touch panel. Alternatively, if the input unit is a mouse, the user may input information by operating the mouse. Alternatively, if the input unit is a keyboard, the user may input information by pressing keys on the keyboard. Alternatively, if the input unit is a microphone, the information may be input by the user inputting voice into the microphone. Alternatively, if the input unit is a camera, information captured by the camera may be input. Alternatively, if the input unit is a data reading device, the information may be input by reading the information from a storage medium connected to the system 100. Alternatively, if the input unit is a receiver, the information may be input by the receiver receiving information from outside the system 100 via a network. In this case, the type of network does not matter. For example, the receiver may receive information via the Internet or may receive information via a LAN.

For example, the input unit of the interface unit 110 receives input of movement conditions from a user's terminal device. For example, the input unit of the interface unit 110 receives input of movement conditions from the database unit 200.

The interface unit 110 includes, for example, an output unit that allows information to be output from the system 100. It does not matter in what manner the output unit allows information to be output from the system 100. For example, if the output unit is a display screen, the information may be output to the display screen. Alternatively, if the output unit is a speaker, the information may be output by sound from the speaker. Alternatively, if the output unit is a data writing device, the information may be output by writing the information to a storage medium connected to the system 100. Alternatively, if the output unit is a transmitter, the transmitter may output the information by transmitting it to the outside of the system 100 via a network. In this case, the type of network does not matter. For example, the transmitter may send information over the Internet or may send information over a LAN.

For example, the output unit of the interface unit 110 transmits the created movement path to each of the multiple moving bodies. For example, the output unit of the interface unit 110 transmits the created movement path to a controller that controls the multiple moving bodies.

The processor unit 120 executes processing of the system 100 and controls the overall operation of the system 100. The processor unit 120 reads a program stored in the memory unit 130 and executes the program. This allows the system 100 to function as a system that executes desired steps. The processor unit 120 may be implemented by a single processor or by multiple processors.

The memory unit 130 stores programs required to execute the processing of the system 100, data required to execute the programs, and the like. The memory unit 130 includes a program for causing the processor unit 120 to perform processing for creating movement routes for each of a plurality of moving objects (for example, implementing the processing shown in FIGS. 7, 8, and 9, which will be described later). program) may be stored. Here, it does not matter how the program is stored in the memory unit 130. For example, the program may be preinstalled in the memory unit 130. Alternatively, the program may be installed in the memory unit 130 by being downloaded via a network. In this case, the type of network does not matter. Memory section 130 may be implemented by any storage means.

The database unit 200 stores, for example, geographic information. Geographical information may include, for example, information regarding topography, information regarding buildings, and the like. Geographical information may also include information regarding no-fly zones.

The database unit 200 may store, for example, movement conditions input by the user, or may store movement routes created in the past, for example.

In the example shown in FIG. 2, the database unit 200 is provided outside the system 100, but the present invention is not limited thereto. It is also possible to provide the database unit 200 inside the system 100. At this time, the database unit 200 may be implemented by the same storage unit that implements the memory unit 130, or may be implemented by a different storage unit from the storage unit that implements the memory unit 130. In any case, the database unit 200 is configured as a storage unit for the system 100. The configuration of the database unit 200 is not limited to a specific hardware configuration. For example, the database unit 200 may be composed of a single hardware component or a plurality of hardware components. For example, the database unit 200 may be configured as an external hard disk device of the system 100, or may be configured as a storage on a cloud connected via a network.

Figure 3 shows an example of the configuration of the processor unit 120.

The processor unit 120 includes an acquisition means 121, a determination means 122, and a derivation means 123.

The acquisition means 121 is configured to acquire information from outside the system 100 via the interface unit 110.

The acquisition means 121 is configured to acquire travel conditions. The movement conditions include a starting point and a plurality of candidate destination points for each of the plurality of moving objects. The starting point and the destination point may be expressed in (latitude, longitude, altitude) coordinates, for example. For example, the movement conditions include a first starting point and a first plurality of candidate arrival points for a first moving object among the plurality of moving objects, and a second moving object among the plurality of moving objects. and a second plurality of candidate destination points.

The movement conditions are, for example,
・Number of moving objects ・Length of one side of the movement range ・A range on the map for route planning ・The longest distance to travel in a straight line (i.e., the length of the straight line route that can be determined)
・Number of points in the point cloud placed around the straight path ・Sampling distance of the obstacle point cloud (i.e., when an obstacle is considered to be a set of multiple obstacle points, Sampling interval. For example, the sampling distance of the obstacle point group can be the length of one side of the grid in the case of grid sampling (sampling one point belonging to the grid).)
・Sampling distance for the path (i.e., the distance between each point when sampling multiple points from each straight path when calculating the repulsive force field)
- Moving object radius (for example, if the moving object is not circular or spherical, this refers to the distance from the center of the moving object to the outermost point of the moving object.)
- It may further include at least one of a stable separation distance between the obstacle and the moving object.

The movement conditions may be input by, for example, a user. Here, the method of input is not important. For example, the user may directly input the movement conditions to the system 100 via the interface unit 110, or the user may input the movement conditions to a terminal device. The movement conditions input to the terminal device may be input to the system 100 via a network, for example. The acquisition means 121 may acquire the input movement conditions.

The movement conditions may be stored in the database unit 200, for example. The movement conditions may be stored in the database unit 200 in advance by the user, for example. The acquisition means 121 can acquire the movement conditions from the database section 200.

The determining means 122 is configured to determine a respective linear route for each of the plurality of moving objects.

The determining means 122 selects one of the plurality of candidate arrival points for each of the plurality of moving objects, and connects the starting point and the selected candidate arrival point of the plurality of candidate arrival points. Straight-line paths for each of a plurality of moving objects can be determined. The selected candidate destination is also referred to as a "selected candidate destination." For example, the determining means 122 determines, for the first mobile object, by connecting the first starting point and one of the first selected candidate arrival points of the first plurality of candidate arrival points. A first linear path for the mobile object may be determined, and for the second mobile object, a second starting point and a second selected candidate destination of one of the second plurality of candidate destinations. By connecting , the second straight path of the second moving body can be determined. At this time, it is preferable that the determining means 122 determine each of the first straight route and the second straight route so that the intersection between the first straight route and the second straight route is minimized. The first linear path and the second linear path may intersect in a three-dimensional space or may intersect on a two-dimensional plane. Preferably, the determining means 122 determines each of the first straight line path and the second straight line path so that the intersection between the first straight line path and the second straight line path on a two-dimensional plane is minimized. do. For example, when two moving objects move along two linear paths that do not intersect in three-dimensional space but intersect when projected onto a two-dimensional plane, a downward flow/upward flow occurs or an aircraft abnormality occurs. If a sudden drop/surge occurs due to this, there is a possibility that the moving objects will collide with each other. By minimizing the intersection on the two-dimensional plane, even this possibility can be eliminated or reduced. Note that the starting point and candidate destination points can each be points in a three-dimensional space, but by projecting them onto a two-dimensional plane (for example, an xy plane), the starting point and candidate destination points can be expressed on a two-dimensional plane. can be handled.

The determination means 122 can determine the straight line path using any algorithm.

In one embodiment, the determining means 122 may determine the straight path using an algorithm that utilizes bipartite graphs. In an algorithm that uses a bipartite graph, each starting point of a plurality of moving objects is taken as a starting point group, multiple candidate arrival points of each of plural moving objects are taken as a destination point group, and one of the starting point groups and the ending point group are Paired with one of them. The determining means 122 determines a straight line connecting one of the starting point groups and one of the ending point groups, and a straight line connecting another one of the starting points and another one of the ending point groups. Pair each of the starting points with one of the ending points such that the intersection of is minimized. That is, the determining means 122 solves a minimum intersection problem. One of the paired starting points and one of the paired ending points correspond to a starting point and a selected candidate destination point, respectively, and a straight line connecting them corresponds to a straight path.

The determining means 122 uses a pruned depth-first search method that counts the number of pair intersections as an algorithm that uses a bipartite graph. The depth-first search method with pruning that counts the number of pair intersections is a method developed by the present inventor. The pruned depth-first search method that counts the number of pair intersections is computationally efficient compared to the method that verifies one of the starting points and one of the ending points one by one to find the pair that has the minimum intersection. It is useful in that it can significantly reduce the amount of data and perform calculations at high speed.

The depth-first search method with pruning that counts the number of pair intersections is a method for searching for pairs of each of the start points and one of the end points so that the number of intersections between a line connecting one of the start points and one of the end points is minimized, by treating multiple pairs whose number of intersections has already been determined as a group and performing a depth-first search (DFS) using (pruning) the calculation results already obtained for the group.

In pruning, pairs whose number of intersections has already been determined are stored in a list along with their number of intersections, and the number of intersections of the pairs in the list can be utilized in subsequent calculations. For example, a first pair of a first point in the starting point group and a first point in the ending point group, and a first pair of a second point in the starting point group and a second point in the ending point group. If it is determined that the number of intersections is 1, the pair of 2 is stored in the list as the number of intersections {(first point, first point), (second point, second point)}=1. In subsequent calculations, the first pair of the first point of the starting point group and the first point of the ending point group, and the second point of the starting point group and the second point of the ending point group. There is no need to recalculate the number of intersections of the second pair with , and the number of intersections 1 obtained from the list can be used instead. This allows the amount of calculation to be reduced.

In the depth-first search method with pruning that counts the number of pair intersections, the following process is repeated in a loop.
- Select one combination of pair candidates from among a plurality of pair candidates between a point in the starting point group and a point in the ending point group using the Priority Queue. A combination includes two or more pair candidates. Priority Queue is an algorithm set to extract combinations that include the most pair candidates and have the least number of intersections.
・Mark the retrieved combination as "reached".
- If the total number of pair candidates included in the extracted combination is the same as the number of starting points (that is, the number of moving objects), the loop ends.
・Mark the number of intersections as reached, assuming that the retrieved subset of combinations is ``at least not greater than the current number of intersections'' (it is already included in the list, and the number of intersections is less than the number of intersections in the list). (If not, do not update).
●Enumerate all possible pairs consisting of points in the starting point group and points in the ending point group that are not included in the combination, and measure the number of intersections by comparing them with the combination.
- At that time, one of the pairs in the combination and one of the listed pairs are selected and their intersection judgment is performed. At this time, if the number of intersections between the selected pairs has already been calculated, it is extracted from the list and the number of intersections is added. If it has not been calculated yet, if it intersects, increase the number of intersections by +1, if it does not intersect, leave it as is and insert the information into the list.

The final combination is the combination of pairs with the minimum intersection.

A pruned depth-first search method that counts the number of pair intersections will be described using a more specific example shown in FIG. 4A.

As shown in FIG. 4A(a), for a starting point group including five starting points and an ending point group including five candidate destination points, one of the starting points and one of the ending point group are connected. Each of the starting points and one of the ending points such that the intersection between the straight line and each straight line connecting another one of the starting points and another one of the ending points is minimized. Consider pairing up.

First, the number of intersections between the pair (1,1) of the first starting point and the pair (2,2) of the second starting point and the second starting point is calculated. In the example shown in FIG. 4A, these pairs do not intersect, so
Intersection number {(1,1), (2,2)} = 0
are stored in the list as

Next, a pair (1,1) of a first starting point and a first starting point, a pair (2,2) of a second starting point and a second starting point, and a third starting point The number of intersections between the pair (3, 3) and the third starting point is calculated. In the example shown in Figure 4A, these pairs do not intersect, so
Number of intersections {(1, 1), (3, 3)} = 0
Number of intersections {(2, 2), (3, 3)} = 0
Number of intersections {(1, 1), (2, 2), (3, 3)} = 0
is stored in the list as . At this time, since {(1, 1), (2, 2)} = 0 has already been calculated and stored in the list, the pair (1, 2) of the first starting point and the first starting point is There is no need to recalculate the number of intersections between 1) and the pair (2, 2) of the second starting point and the second starting point, and the pair (2, 2) of the first starting point and the first starting point 1,1) and the pair (3,3) of the third starting point and the third starting point, and the number of intersections of the pair (2,2) of the second starting point and the second starting point ) and the pair (3, 3) of the third starting point and the third starting point. By performing pruning in this way, the amount of calculation can be significantly reduced.

Next, a pair (1,1) of a first starting point and a first starting point, a pair (2,2) of a second starting point and a second starting point, and a third starting point The number of intersections between the pair (3, 3) of the and the third starting point, and the pair (4, 4) of the fourth starting point and the fourth starting point are calculated. In the example shown in Figure 4A, these pairs do not intersect, so
Number of intersections {(1, 1), (4, 4)} = 0
Number of intersections {(2, 2), (4, 4)} = 0
Number of intersections {(3, 3), (4, 4)} = 0
Number of intersections {(1, 1), (2, 2), (4, 4)} = 0
Number of intersections {(1, 1), (3, 3), (4, 4)} = 0
Number of intersections {(2, 2), (3, 3), (4, 4)} = 0
Number of intersections {(1, 1), (2, 2), (3, 3), (4, 4)} = 0
is stored in the list as . At this time, since {(1, 1), (2, 2), (3, 3)} = 0 has already been calculated and stored in the list, the first starting point and the first starting point the pair (1,1), the pair (2,2) of the second starting point and the second starting point, and the pair (3,3) of the third starting point and the third starting point. There is no need to recalculate the number of intersections between the first starting point and the first starting point (1,1), and the fourth starting point and the fourth starting point pair (4, 4), the number of intersections between the pair (2,2) of the second starting point and the second starting point and the pair (4,4) of the fourth starting point and the fourth starting point and the number of intersections between the pair (3,3) of the third starting point and the third starting point and the pair (4,4) of the fourth starting point and the fourth starting point. That's enough.

Next, as shown in FIG. 4A(b), the pair (1,1) of the first starting point and the first starting point and the pair (1,1) of the second starting point and the second starting point 2,2), the pair (3,3) of the third starting point and the third starting point, the pair (4,4) of the fourth starting point and the fourth starting point, and the fifth starting point Calculate the number of intersections between the point and the pair (5,5) of the fifth starting point. In the example shown in FIG. 4A(b), these pairs have one intersection, so
Number of intersections {(1, 1), (2, 2), (3, 3), (4, 4), (5, 5)} = 1
is stored in the list as . At this time, since {(1,1), (2,2), (3,3), (4,4)}=0 has already been calculated and stored in the list, the first starting point and the first starting point (1,1), the second starting point and the second starting point (2,2), and the third starting point and the third starting point. There is no need to recalculate the number of intersections between the pair (3, 3) and the pair (4, 4) between the fourth starting point and the fourth starting point; The number of intersections between the pair (1, 1) and the pair (5, 5) between the fifth starting point and the fifth starting point, and the pair (2 , 2), the number of intersections between the pair (5, 5) of the fifth starting point and the fifth starting point, and the pair (3, 3) of the third starting point and the third starting point, The number of intersections between the pair (5, 5) of the fifth starting point and the fifth starting point, and the number of intersections between the pair (4, 4) of the fourth starting point and the fourth starting point and the fifth starting point It is sufficient to calculate the number of intersections between the starting point and the fifth starting point pair (5, 5).

Next, from the perspective of depth-first search, as shown in FIG. 4A(c), the pair (1,1) of the first starting point and the second starting point and the The pair (2, 2) of the second starting point, the pair (3, 3) of the third starting point and the third starting point, and the pair (4, 3) of the fourth starting point and the fifth starting point. , 5), calculate the number of intersections of the pair (5, 4) of the fifth starting point and the fourth starting point. In the example shown in FIG. 4A(c), these pairs do not intersect, so
Number of intersections {(1, 1), (2, 2), (3, 3), (4, 5), (5, 4)} = 0
is stored in the list as . At this time, since {(1, 1), (2, 2), (3, 3)} = 0 has already been calculated and stored in the list, the first starting point and the first starting point the pair (1,1), the pair (2,2) of the second starting point and the second starting point, and the pair (3,3) of the third starting point and the third starting point. There is no need to recalculate the number of intersections between the first starting point and the first starting point (1,1), and the fourth starting point and the fifth starting point pair (4, 5), the number of intersections between the pair (2,2) of the second starting point and the second starting point and the pair (4,5) of the fourth starting point and the fifth starting point number, the number of intersections between the pair (3,3) of the third starting point and the third starting point and the pair (4,5) of the fourth starting point and the fifth starting point, the first starting point The number of intersections between the pair (1, 1) of the point and the first starting point and the pair (5, 4) of the fifth starting point and the fourth starting point, and the number of intersections between the second starting point and the second starting point The number of intersections between the pair (2, 2) with the starting point and the pair (5, 4) between the fifth starting point and the fourth starting point, and the pair between the third starting point and the third starting point (3,3) and the pair (5,4) of the fifth starting point and the fourth starting point, and the number of intersections of the pair (4,5) of the fourth starting point and the fifth starting point. ) and the pair (5, 4) of the fifth starting point and the fourth starting point.

Continue depth-first search for the same number of five pairs as starting points until the number of intersections is determined to be the minimum. For example, if it is determined that the number of intersections will not become smaller than the currently obtained minimum number of intersections even after further searching, the process can be terminated. In this example, since the minimum number of intersections is 0, the process can be ended here.

In an algorithm using a bipartite graph, the determining means 122 may use a sequential partition method to divide the minimum intersection problem into subproblems before solving the minimum intersection problem. The Sequential Partition method is a method developed by the present inventor. By dividing the minimum intersection problem into subproblems, the amount of calculation when solving the minimum intersection problem can be significantly reduced. When solving a minimum intersection problem for all points, for example, if a pruned depth-first search method that counts the number of pair intersections is used, the amount of calculation can be O(N ⁴ ), but the minimum intersection problem This is because N can be reduced by solving a subproblem of , that is, by solving a minimum intersection problem for each of a plurality of subsets of the point cloud (N is the number of starting points). In one embodiment, when the number of moving objects is 15, it takes more than one day to calculate the minimum intersection problem for all points, but when using the sequential partition method, it takes about 20 minutes. can do.

FIG. 4B schematically illustrates the steps of the Sequential Partition method.

As shown in FIG. 4B(a), it is assumed that a two-dimensional distribution of starting points for each of the plurality of moving bodies and a plurality of candidate arrival points for each of the plurality of moving bodies is given. Although the starting point and the candidate destination points can each be points in a three-dimensional space, the two-dimensional distribution shown in FIG. 4B(a) is a distribution in which the starting point and the candidate destination points are projected onto the xy plane.

Next, as shown in FIG. 4B(b), the axis of the two-dimensional distribution shown in FIG. 4B(a) is rotated, and the distribution is converted into a horizontally long distribution in the x-axis direction. This can be done, for example, by any multivariate analysis, for example by performing KL expansion on the two-dimensional distribution (calculating a subspace that best approximates the vector distribution). At this time, the x-axis corresponds to the first dimension of the eigenvalues, and the y-axis corresponds to the second dimension of the eigenvalues.

Next, one of the plurality of starting points is selected, and one of the plurality of candidate destination points is temporarily placed as the pair. The selected starting point can be any point among the plurality of starting points. Based on the straight line obtained by connecting the selected starting point and the paired candidate destination point, it is determined whether there are the same number of starting points and candidate destination points on the negative side or the positive side, and the same number of starting points and candidate destination points exist. If a point and a candidate arrival point exist, the pair can become a Partition, and the negative and positive sides of the pair can be separated and handled separately. This is because the number of starting points that exist on the negative or positive side is the same as the number of candidate destination points, so it is possible to form pairs only with starting points and candidate destination points that exist on the negative or positive side. Because it can be done. A pair of straight lines that can become a Partition is a straight line that never intersects with any other pair of straight lines.

In a certain case, as shown in FIG. 4B(c), the leftmost starting point S1 (i.e., the x-axis value is the smallest) is selected, and the leftmost candidate destination point G1 (i.e., the x-axis value is the smallest) is provisionally set as its pair. In this case, with respect to the line L11 obtained by connecting S1 and G1, if the number of starting points and the number of candidate destination points on the negative side of L11 are the same (or the number of starting points and the number of candidate destination points on the positive side of L11 are the same), this pair (S1, G1) can be a Partition, so the pair (S1, G1) is determined. In the example shown in FIG. 4B(c), the number of starting points and the number of candidate destination points on the positive side are the same, at 8.

In another case, as shown in FIG. 4B(d), select the third starting point S3 from the left (i.e., the third smallest x-axis value) and the fourth starting point S3 from the left (i.e., x Candidate arrival point G4 (with the fourth smallest axis value) is temporarily placed as the pair. At this time, with respect to the straight line L34 obtained by connecting S3 and G4, the number of starting points that exist on the negative side of L34 and the number of candidate destination points are the same (or the number of starting points that exist on the positive side of L34). If the number of starting points and the number of candidate destination points are not the same), this pair (S3, G4) cannot be a Partition, and the search moves on to another pair. In the example shown in FIG. 4B(c), the number of starting points on the positive side and the number of candidate arrival points are 6 and 5, respectively, which do not match, and the number of starting points and candidates on the negative side do not match. The number of reached points is 2 and 3, respectively, which do not match. To select another pair, for example, temporarily place the candidate destination point with the next smallest or largest x-axis value as a pair, or select the starting point with the next smallest or largest x-axis value. etc.

Note that if there are multiple starting points and/or candidate arriving points on the straight line obtained by connecting the selected starting point and the paired candidate arriving points, the pair with the minimum distance is selected as Partition. It can be adopted as

When the number of starting points is N, the amount of calculation by the Sequential Partition method is O(N 2 ) in the maximum case (when all are not paired), and O(N ² ) in the minimum case (when all are paired at once). ), O(N).

After applying the Sequential Partition method, for example, a minimum intersection problem is solved for the subset of starting points and candidate destination points carved out by the Sequential Partition method. For example, the minimum intersection problem for a subset can be solved by the pruned depth-first search method that counts the number of pair intersections described above. Alternatively, after applying the Sequential Partition method, for example, the subsets may be divided by further applying the Sequential Partition method to the subsets of the starting point and candidate destination points that have been divided by the Sequential Partition method. This is because by reducing the number of starting points in the subset, the amount of calculation required to solve the minimum intersection problem can be reduced. Preferably, the Sequential Partition method should be applied such that the amount of calculation for the Sequential Partition method does not exceed the amount of calculation for solving the minimum intersection problem, which can be reduced by the Sequential Partition method.

The deriving means 123 is configured to derive a corrected route based on each straight line route determined for each of the plurality of moving objects by the determining means 122.

The derivation means 123 specifies at least one point between the starting point and the selected candidate destination point for each linear route of the plurality of moving objects, and identifies the starting point, the specified at least one point, and the selected candidate. By correcting the straight line route by connecting the destination points, corrected routes for each of the plurality of moving bodies can be derived. For example, the deriving means 123 calculates, for the first straight route, a first starting point, at least one point between the first starting point and the first selected candidate destination, and a first selected candidate destination point. By modifying the first linear route by connecting the points, the first modified route can be derived. For example, the deriving means 123 calculates, for the second linear route, the second starting point, at least one point between the second starting point and the second selected candidate destination, and the second selected candidate destination point. By modifying the second linear route by connecting the points, a second modified route can be derived.

At least one point between the starting point and the selection candidate destination point may be at least one point of a group of points arranged around each of the multiple linear paths. For example, at least one point between the first starting point and the first selection candidate destination point may be at least one point of a first group of points arranged around the first linear path, and at least one point between the second starting point and the second selection candidate destination point may be at least one point of a second group of points arranged around the second linear path.

Here, the point cloud may have an arbitrary shape of extent around each of the plurality of linear paths. For example, the point cloud may have a spherical extent centered at the starting point, the selected candidate destination, or a point between the starting point and the selected candidate destination, but may have other shapes, such as ellipsoidal, spherical, It may have a cubic, rectangular parallelepiped, cylindrical, or prismatic spread. Further, the point cloud may have an arbitrarily large extent around each of the plurality of straight-line paths. For example, the point cloud has an extent where the distance between the starting point and the selected candidate destination is the length of a diameter, the length of a radius, the length of a side, or the length of a diagonal. For example, if the distance between the starting point and the candidate destination point is approximately 10 km, it is preferable that the point group has a spherical extent with a radius of approximately 10 km.

Furthermore, a point cloud can have any number of points. For example, the point cloud may have a number of points depending on the distance between the starting point and the selected candidate destination. For example, a point cloud can have up to about 100 points, up to about 500 points, up to about 1000 points, up to about 2000 points, up to about 10000 points. Increasing the number of points in the point cloud enables more precise route generation, but increases the amount of calculation. On the other hand, reducing the number of points in the point cloud allows for faster calculations, but may result in crude path generation. Preferably, it may be a point cloud having about 500 to about 1000 points. This is because there is a good balance between the amount of calculation and the precision of the route.

Furthermore, the point cloud may have any point distribution. For example, a point cloud may have a random point distribution. That is, the points within the point cloud may be randomly distributed within the space occupied by the point cloud. For example, the point cloud may have a uniform point distribution. That is, the points within the point cloud may be regularly distributed within the space occupied by the point cloud. In a preferred embodiment, the point cloud may have a point distribution such that the points become denser closer to the starting point, or a point distribution such that the points become denser closer to the selected candidate destination. In a further preferred embodiment, the point distribution may be such that the closer the starting point is, the more dense the points are, and the closer the selected candidate destination point is, the denser the points are. According to a point cloud having such a distribution, a larger number of points are provided near the starting point and/or the selected candidate destination point, and a smaller number of points are provided farther from the starting point and/or the selected candidate destination point. I can do it. This enables more precise route generation by making it possible to finely modify the straight line route near the starting point and/or selected candidate destination point where multiple moving objects are likely to collide. , it is possible to reduce the number of options for distant starting points and/or selection candidate destination points where there is a low possibility that multiple moving bodies will collide, thereby reducing the calculation load.

A point cloud having a point distribution in which the points are denser closer to the starting point is placed around the linear path, for example, by the method shown in FIG. 5A. In the example shown in FIG. 5A, for ease of description, the point cloud is shown only in the first quadrant in a two-dimensional plane, but it is understood that the point cloud may naturally be distributed in the second, third, and/or fourth quadrants as well as in three-dimensional space.

First, as shown in FIG. 5A(a), a point group having a random distribution is obtained in a polar coordinate system.

Second, as shown in FIG. 5A(b), the point group in the polar coordinate system is transformed into the point group in the orthogonal coordinate system. As a result, the point group has a point distribution in which the closer the point is to the origin, the denser the points, and the farther away from the origin, the sparser the points. This point group has a circular (spherical in three-dimensional space) spread.

For example, the coordinate p _i of point i in the point group in a two-dimensional polar coordinate system is
p _i = (d _i , θ _i ) (d: (0, 1], θ: [0, 2π))
When a point group in a two-dimensional polar coordinate system is converted to a point group in a two-dimensional orthogonal coordinate system, the coordinates q _i of point i in the point group in the two-dimensional orthogonal coordinate system are
q _i =(x _i , y _i , z _i )=(d _i cos(θ _i )), d _i sin(θ _i ))
becomes.

For example, the coordinate p _i of point i in the point group in a three-dimensional polar coordinate system is
p _i = (d _i , θ _i , φ _i ) d: (0, 1], θ: [0, 2π), φ: [-1/2π, 1/2π]
When a point group in a three-dimensional polar coordinate system is converted to a point group in a three-dimensional orthogonal coordinate system, the coordinates q _i of point i in the point group in the three-dimensional orthogonal coordinate system are
q _i = (x _i , y _i , z _i ) = (d _i sin (θ _i ) cos (φ _i ), d _i sin (θ _i ) sin (φ _i ), d _i cos (θ _i ))
becomes.

Thirdly, as shown in FIG. 5A(c), the point cloud in the Cartesian coordinate system is positioned with the starting point at its center. This makes it possible to obtain a point cloud having a point distribution in which the points become denser as they approach the starting point. At this time, the spread of the point cloud can be adjusted so that the point cloud extends to the selection candidate destination point.

A point group having a point distribution such that the points become denser as they approach the selected candidate destination point is arranged around the straight path by, for example, a method similar to the method shown in FIG. 5A. In the method shown in FIG. 5A, by arranging the point group in the orthogonal coordinate system around the selected candidate destination point, it is possible to obtain a point group with a point distribution in which the points become denser as they approach the selected candidate destination point. can.

A point group having a point distribution such that the closer to the starting point the closer the points are and the closer the selected candidate destination point the more dense the points are, for example, arranged around the straight path by the method shown in FIG. 5B. be done.

First, as shown in FIG. 5B(a), a point group having a random distribution is obtained in a polar coordinate system.

Second, as shown in FIG. 5B(b), the point group in the polar coordinate system is transformed into the point group in the orthogonal coordinate system. As a result, the point group has a point distribution in which the closer the point is to the origin, the denser the points, and the farther away from the origin, the sparser the points. This point group has a circular (spherical in three-dimensional space) spread.

Third, as shown in FIG. 5B(c), a group of points in the orthogonal coordinate system is arranged around the starting point, and the same group of points is arranged around the selected candidate arrival point. As a result, it is possible to obtain a point group having a point distribution such that the closer the starting point is, the more dense the points are, and the closer the selected candidate destination point is, the denser the points are. At this time, the spread of the point group can be adjusted so that the point group covers the entire area between the starting point and the selected candidate destination point. In this distribution, the distance between the starting point and the selected candidate destination point is sparser than the vicinity of the starting point and the selected candidate destination point.

The deriving means 123 can identify at least one point between the starting point and the selected candidate destination point for each straight path of the plurality of moving objects, taking obstacles into consideration. For example, the deriving means 123 can calculate the repulsive force field that an obstacle exerts on the straight path, and specify at least one point between the starting point and the selected candidate destination point based on the repulsive force field. The deriving means 123 specifies at least one point between the starting point and the selected candidate destination point, for example, so that the repulsive force field is minimized or the repulsive force field is smaller than a predetermined threshold value. I can do it. The deriving means 123 may, for example, identify at least one point between the starting point and the selected candidate destination point so that the repulsive force field is smaller and the length of the resulting path is shorter. can. The deriving means 123 calculates, for example, such that the sum of the value of the repulsive force field and the length of the resulting path is the minimum, or the value of the repulsive force field and the length of the resulting path are At least one point between the starting point and the selected candidate destination point can be identified such that the sum of the values of the starting point and the selected candidate destination point is smaller than a predetermined threshold value. Obstacles can be, for example, terrain (eg, mountains, hills, trees, etc.), buildings, or the like.

The repulsive field here can indicate the risk of collision with an obstacle. The closer you get to the obstacle, the stronger the repulsive field becomes, and therefore the higher the risk of collision.

The repulsive field that an obstacle exerts on a straight line path can be calculated, for example, by regarding the obstacle as a set of multiple obstacle points and calculating the repulsive field that the obstacle points in the vicinity of the straight line path exert on the straight line path. The multiple obstacle points are sampled, for example, by lattice sampling (sampling one point on the obstacle for each lattice). The multiple obstacle points can be sampled at any sampling distance. The multiple points of the obstacle are stored in the database unit 200, for example, by using a KD tree. This allows the derivation means 123 to search for the nearest obstacle point from each of the multiple points on the straight line path, for example, by using the KD tree. The multiple points on the straight line path can be, for example, any number of arbitrary points, for example, any number of points that divide the straight line path at equal intervals. The multiple points on the straight line path can be sampled at any sampling distance.

ここで、Ｏ_ｗは、障害物からの離隔距離であり、それ以上障害物に近づくと危険であることを示す距離である。Ｏ_ｗは、障害物の性質等に応じて任意の値に設定され得る。Ｏ_ｓは、移動体の半径である（例えば移動体が円形または球形でない場合、移動体の中心から移動体の最外点までの距離をいう）。ｐ_ｉは、直線経路上のｉ番目の点の座標を表し、ｏ_ｉは、ｐ_ｉから一番近い障害物点の座標を表す。ｄｉｓｔ（ｐ_ｉ－ｏ_ｉ）は、ｐ_ｉとｏ_ｉとの間の距離を表す。なお、ｄｉｓｔ（ｐｉ－ｏｉ）が０以下であるとき（すなわち、直線経路と障害物とが衝突する場合）は、ｄｉｓｔ（ｐｉ－ｏｉ）を極小値（例えば、０．００００１に設定することができる。例えば、移動体の最外部がちょうど障害物からの離隔距離上にある場合には、ｄｉｓｔ（ｐｉ－ｏｉ）－Ｏ_ｓ＝Ｏ_ｗとなることから、その点ｉでの斥力場ｗ_ｐｉは、２となる。移動体が離隔距離よりも障害物から離れると斥力場は２よりも小さくなり、移動体が離隔距離よりも障害物に近づくと斥力場は２よりも大きくなる。 The derivation means 123 can calculate the repulsive field w _p for each of the multiple points on the straight line path using the following equation, for example.

Here, O _w is the distance from the obstacle, and is the distance indicating that it is dangerous to approach the obstacle any closer. _{O w} can be set to any value depending on the nature of the obstacle, etc. _{O s} is the radius of the moving object (for example, if the moving object is not circular or spherical, it refers to the distance from the center of the moving object to the outermost point of the moving object). p _i represents the coordinates of the i-th point on the straight path, and o _i represents the coordinates of the obstacle point closest to p _i . dist(p _i -o _i ) represents the distance between p _i and o _i . When dist(pi-oi) is 0 or less (i.e., when the straight path collides with an obstacle), dist(pi-oi) can be set to a minimum value (e.g., 0.00001. For example, when the outermost part of the moving body is exactly at the separation distance from the obstacle, dist(pi-oi) _-Os = _Ow , and therefore the repulsive field _wpi at point i is 2. When the moving body is farther away from the obstacle than the separation distance, the repulsive field becomes smaller than 2, and when the moving body is closer to the obstacle than the separation distance, the repulsive field becomes larger than 2.

Note that the equation of Equation 1 is just an example, and the repulsive force field can be derived using other equations. For example, the coefficient 2 of _Ow can be changed to any value.

The deriving means 123 can calculate the repulsive force field w _p of the straight line path by summing up the repulsive force fields w _{p i} calculated for a plurality of points on the straight line path.

After calculating the repulsive field, the derivation means 123 can use, for example, the A* algorithm to identify at least one point between the starting point and the selected candidate destination point such that the repulsive field is minimized. Alternatively, after calculating the repulsive field, the derivation means 123 can use, for example, the A* algorithm to identify at least one point between the starting point and the selected candidate destination point such that the sum of the repulsive field value and the resultant path length value is minimized. For example, the further away from the obstacle, the smaller the repulsive field becomes, but the longer the resultant path length becomes. In the A* algorithm, it is possible to search for a point where the sum of the repulsive field value w _p and the resultant path length value h is minimum.

For example, the deriving means 123 calculates, for the first straight path, (a) the repulsive force field that an obstacle exerts on the first straight path, and (b) calculates the first departure so that the repulsive force field is minimized. identifying at least one point between the point and the first selection candidate destination; (c) the first starting point, the first selection candidate destination, the first starting point and the first selection; The first linear path may be modified by connecting at least one point between the candidate destination and the candidate destination. Thereby, the derived first modified route can be a route in which the possibility of colliding with an obstacle is minimized.

For example, the deriving means 123 calculates, for the first straight path, (a) the repulsive force field that an obstacle gives to the first straight path, and (b) the repulsive force field and the length of the resulting path. (c) identify at least one point between the first starting point and the first selection candidate arrival point such that the sum of the values of the first starting point and the first The first straight path can be modified by connecting the selected candidate destination point of the first selected candidate destination point and at least one point between the first starting point and the first selected candidate destination point. As a result, the derived first modified route can be a route with a reduced possibility of colliding with an obstacle and a reduced route length.

In addition to considering as an obstacle an actual obstacle that may exist between the starting point and the selected candidate destination point, the deriving means 123 considers the modified route already derived by the deriving means 123 as an obstacle. Then, a modified route may be derived by modifying another straight route. For example, with respect to the second straight path, the deriving means 123 (a) calculates a repulsive force field that an obstacle exerts on the second straight path, and (b) calculates a second starting point and a second starting point based on the repulsive force field. (c) the second starting point, the second selected candidate destination, and the second starting point and the second selected candidate destination; After the second straight path is corrected by connecting at least one point between (a) calculating the repulsive force field exerted by the obstacle on the first straight path; and (b) calculating at least the distance between the first starting point and the first selected candidate destination point based on the repulsive force field. identifying a point; (c) connecting the first starting point, the first selected candidate destination, and at least one point between the first starting point and the first selected candidate destination; Accordingly, the first straight path can be corrected and the first corrected path can be derived. By considering the already derived modified route as an obstacle, not only is the possibility of collision with an actual obstacle that may exist between the starting point and the selected candidate destination point minimized, but also other It becomes possible to create a route in which the possibility of colliding with the route of a moving object is minimized.

For example, for a modified route that is initially derived (i.e., a modified route that does not consider the already derived modified route as an obstacle), after another modified route is derived, Taking this into consideration, a modified route may be derived from the straight route again.

6A to 6E schematically show a flow of identifying at least one point between a starting point and a selected candidate destination point using the A* algorithm and deriving a modified route. In the examples shown in FIGS. 6A to 6E, n starting points S ₁ to S _n and n corresponding selection candidate destination points G ₁ to G _n are shown, and S ₁ and G ₁ are , between S ₂ and G ₂ , between S ₃ and G ₃ , . . . between S _n and G _n . In the examples shown in FIGS. 6A to 6E, actual obstacles are omitted to simplify the explanation, and each straight path is set as an obstacle. Furthermore, in the examples shown in FIGS. 6A to 6E, the starting points S ₁ to S _n , the selection candidate destination points G ₁ to G _n , and the straight line are depicted two-dimensionally to simplify the explanation. It will be appreciated that the path, and the modified path, may be located in three-dimensional space.

In FIG. 6A, one starting point out of n starting points is randomly selected. Here, the third starting point S ₃ has been selected, and the third straight path connecting S ₃ and G ₃ will be modified.

In FIG. 6B, the third straight path is erased and a point cloud is placed between S ₃ and G ₃ . A point cloud is any point cloud acquired and arranged as described above.

In FIG. 6C, A* is calculated for each path that follows at least one of the point clouds. The weighting of each route used when calculating A* is the repulsive force field _wp given by each obstacle when other straight routes are obstacles and the heuristic function h. Here, the heuristic function h is a term that acts to suppress deviation from a straight path. A simple example may be expressed as h=-kx, where k is a positive constant and x is the deviation distance from the straight path.

A* is calculated so that the repulsive field _wp is smaller and the length of the resulting path is shorter. For example, in Fig. 6C, the repulsive field _wp and the heuristic function h are calculated for the paths shown by the dotted and double-dashed lines, and the path where the repulsive field _wp is smaller and the heuristic function h is smaller is shown by the double-dashed line. This identifies points in the point cloud that form a path where the repulsive field _wp is smaller and the heuristic function h is smaller.

In FIG. 6D, S ₃ is connected to a point in the point cloud that forms a path where the repulsive force field w _p is smaller and the heuristic function h is smaller, and G ₃ is connected to the point connecting S ₃ and G ₃ . Three modified paths have been derived.

Then, one starting point among the n starting points is randomly selected and the procedure shown in FIGS. 6A-6D is repeated. At that time, for example, the already derived modified route is also set as an obstacle.

FIG. 6E shows the first to nth modified paths derived as a result of repeating the procedure shown in FIGS. 6A to 6D for all starting points.

The corrected routes for each of the plurality of moving objects derived by the derivation means 123 are outputted to the outside of the system 100 as the respective movement routes for the plurality of moving objects. The movement route is transmitted to each of the plurality of moving objects, for example, via the interface unit 110. Alternatively, the movement route is transmitted, for example, via the interface unit 110 to a controller that controls a plurality of moving objects. By following their respective movement paths, a plurality of moving objects can move while minimizing the possibility of colliding with obstacles and other moving objects.

In the example shown in FIG. 3 described above, each component of the processor section 120 is provided within the same processor section 120, but the present invention is not limited to this. A configuration in which each component of the processor unit 120 is distributed among multiple processor units is also within the scope of the present invention. In this case, the plurality of processor units may be located within the same hardware component, or may be located within separate nearby or remote hardware components.

Note that each component of the system 100 described above may be composed of a single hardware component or may be composed of a plurality of hardware components. When configured with a plurality of hardware components, it does not matter how each hardware component is connected. Each hardware component may be connected wirelessly or by wire. The system 100 of the present invention is not limited to any particular hardware configuration. It is also within the scope of the present invention to configure the processor section 120 with an analog circuit rather than a digital circuit. The configuration of the system 100 of the present invention is not limited to that described above as long as its functions can be realized.

3. Processing performed by the system for creating travel routes for each of a plurality of moving objects FIG. 7 is a flowchart illustrating an example of a process 700 performed by the system 100 for creating travel routes for each of a plurality of mobile objects. Process 700 may be performed by processor portion 120 of system 100.

In step S701, the acquisition means 121 of the processor unit 120 acquires the travel conditions. The travel conditions include a starting point and a plurality of candidate destination points for each of the plurality of moving bodies. For example, the travel conditions include a first starting point and a first plurality of candidate destination points for a first moving body of the plurality of moving bodies, and a second starting point and a second plurality of candidate destination points for a second moving body of the plurality of moving bodies. The acquired travel conditions are passed to the determination means 122 of the processor unit 120.

In step S702, the determining means 122 of the processor unit 120 determines a straight path for each of the plurality of moving objects. For example, the determining means 122 selects one of the plurality of candidate arrival points for each of the plurality of moving objects, and connects the starting point and the selected candidate arrival point among the plurality of candidate arrival points. Accordingly, each straight path of a plurality of moving objects can be determined. For example, the determining means 122 determines, for the first mobile object, by connecting the first starting point and one of the first selected candidate arrival points of the first plurality of candidate arrival points. A first linear path for the mobile object may be determined, and for the second mobile object, a second starting point and a second selected candidate destination of one of the second plurality of candidate destinations. By connecting , it is possible to determine the second straight path of the second moving body.

The determining means 122 can determine each of the first straight path and the second straight path so that the intersection between the first straight path and the second straight path is minimized. The intersection of the first linear path and the second linear path may be, for example, an intersection on a two-dimensional plane.

The determining means 122 can determine the straight path using any algorithm. For example, the determining means 122 can determine a straight path using an algorithm that utilizes a bipartite graph.

For example, the determining means 122 solves a minimum intersection problem using a depth-first search method with pruning that counts the number of pair intersections, so that the determining means 122 calculates Each of the straight line paths can be determined.

For example, the determining means 122 divides the minimum intersection problem into subproblems using the Sequential Partition method, and then solves the minimum intersection problem for each of the divided subproblems, thereby determining each of the plurality of straight line paths. Each of the plurality of straight-line paths can be determined such that the intersection of the lines is minimized. In one embodiment, a sequential partition method is used to divide the minimum intersection problem into subproblems, and a pruned depth-first search method is used to count the number of pairwise intersections for each of the divided subproblems. By solving the subproblems of the minimum intersection problem, each of the plurality of straight-line routes can be determined so that the intersection of each of the plurality of straight-line routes is minimized. By dividing the minimum intersection problem into subproblems and then solving them, the amount of calculation can be significantly reduced.

In one example, when a certain machine calculates straight-line paths from the starting point to the destination point for 15 moving objects, by using the pruned depth-first search method that counts the number of pair intersections, It took about 1.5 days to complete the calculation until the corrected route was derived using A*. If the same machine performs the Sequential Partition method on 15 moving objects, and then performs the pruned depth-first search method that counts the number of pair intersections, up to the derivation of a modified path using A*, which will be described later The calculation could be completed in about 70 minutes. In this way, by combining the Sequential Partition method and the pruned depth-first search method that counts the number of pair intersections, we were able to complete the calculations at an unexpectedly high speed.

The determined plural straight-line routes are passed to the deriving means 123.

In step S703, the deriving means 123 of the processor unit 120 derives a modified route by modifying at least one of the straight-line routes determined in step S702. The deriving means 123 specifies at least one point between the starting point and the selected candidate destination point for each straight path of the plurality of moving objects, and identifies the starting point, the specified at least one point, and the selected candidate. By correcting the straight line route by connecting the destination points, corrected routes for each of the plurality of moving bodies can be derived. For example, the deriving means 123 calculates, for the first straight path, a first starting point, at least one point between the first starting point and the first selected candidate destination, and a first selected candidate destination point. By modifying the first linear route by connecting the points, the first modified route can be derived.

The at least one point between the starting point and the selected candidate destination point may be at least one point of a group of points arranged around each of the plurality of straight-line paths. For example, the at least one point between the first starting point and the first selected candidate destination point is at least one point of a first point group arranged around the first straight path. and the at least one point between the second starting point and the second selected candidate destination point is at least one point of a second point group arranged around the second straight path. obtain.

The points are arranged around the straight line path, for example, by the process shown in FIG. 8.

FIG. 8 is a flowchart illustrating an example of a process 800 for arranging a point group around a straight line path connecting a starting point and a selected candidate destination point. Process 800 may be performed by processor portion 120 of system 100.

In step S801, the deriving means 123 of the processor unit 120 acquires a point group having a random distribution in the polar coordinate system.

In step S802, the derivation unit 123 of the processor unit 120 converts the point group in the polar coordinate system acquired in step S801 into a point group in the orthogonal coordinate system. As a result, the point group has a point distribution in which the closer the point is to the origin, the denser the points, and the farther away from the origin, the sparser the points.

In step S803, the derivation means 123 of the processor unit 120 centers the point cloud transformed in step S802 around the starting point. The derivation means 123 can center the transformed point cloud for each of the multiple straight line paths around the respective starting point. For example, the derivation means 123 can center the transformed point cloud for the first straight line path around the first starting point.

In step S804, the deriving means 123 of the processor unit 120 arranges the point group converted in step S802 centered on the candidate arrival point. The deriving means 123 can center the transformed point clouds around the respective candidate arrival points for each of the plurality of straight-line paths. For example, the deriving means 123 may center the transformed point cloud on the first candidate arrival point for the first straight path.

The point cloud arranged by process 800 has a point distribution in which the points become denser the closer to the starting point on each of the multiple straight line paths, and the closer to the selected candidate destination point, the denser the points become.

In the above example, it has been described that all steps of process 800 are performed by processor unit 120, but at least one of the steps of process 800 can be performed by a processor unit other than processor unit 120. For example, steps S801 and S802 can be performed in advance by a processor unit other than processor unit 120, and the point cloud in the orthogonal coordinate system can be stored in advance in database unit 200. This allows processor unit 120 to perform steps S803 and S804 using the point cloud stored in database unit 200.

The deriving means 123 can derive a corrected route for each straight route of a plurality of moving objects, for example, by the process shown in FIG.

FIG. 9 is a flowchart illustrating an example of a process 900 for deriving a modified route. Process 900 may be performed by processor portion 120 of system 100. In process 900, a first corrected route is derived for a first linear route among the respective linear routes of a plurality of moving objects.

ここで、Ｏ_ｗは、障害物からの離隔距離であり、それ以上障害物に近づくと危険であることを示す距離である。Ｏ_ｗは、障害物の性質等に応じて任意の値に設定され得る。Ｏ_ｓは、移動体の半径である（例えば移動体が円形または球形でない場合、移動体の中心から移動体の最外点までの距離をいう）。ｐ_ｉは、直線経路上のｉ番目の点の座標を表し、ｏ_ｉは、ｐ_ｉから一番近い障害物点の座標を表す。ｄｉｓｔ（ｐ_ｉ－ｏ_ｉ）は、ｐ_ｉとｏ_ｉとの間の距離を表す。なお、ｄｉｓｔ（ｐｉ－ｏｉ）が０以下であるとき（すなわち、直線経路と障害物とが衝突する場合）は、ｄｉｓｔ（ｐｉ－ｏｉ）を極小値（例えば、０．００００１に設定することができる。 In step S901, the derivation means 123 of the processor unit 120 calculates a repulsive force field that the obstacle exerts on the first straight path. The repulsive force field that the obstacle exerts on the first straight path can be calculated, for example, by regarding the obstacle as a set of multiple obstacle points and calculating the repulsive force field that the obstacle points in the vicinity of the first straight path exert on the first straight path. For example, the derivation means 123 can calculate the repulsive force field _wp for each point of the multiple points on the first straight path using the following formula:

Here, O _w is the distance from the obstacle, and is the distance indicating that it is dangerous to approach the obstacle any closer. _{O w} can be set to any value depending on the nature of the obstacle, etc. _{O s} is the radius of the moving body (for example, if the moving body is not circular or spherical, it refers to the distance from the center of the moving body to the outermost point of the moving body). p _i represents the coordinate of the i-th point on the straight path, and o _i represents the coordinate of the obstacle point closest to p _i . dist(p _i -o _i ) represents the distance between p _i and o _i . Note that when dist(p i -o i ) is 0 or less (i.e., when the straight path and the obstacle collide), dist(p i -o i ) can be set to a minimum value (for example, 0.00001.

The deriving unit 123 can calculate the repulsive force field w _p of the first straight path by summing up the repulsive force fields w _{p i} calculated for a plurality of points on the first straight path.

In step S902, the derivation means 123 of the processor unit 120 connects the first starting point, the first selection candidate destination point, and at least one point based on the repulsive field. The at least one point is identified from the group of points arranged by the process 800 so that the repulsive field is minimized.

The deriving means 123 can identify at least one point based on the repulsive force field using, for example, the A* algorithm. The deriving means 123 can use, for example, the A* algorithm to identify at least one point between the starting point and the selected candidate destination point so that the repulsive force field is minimized. Alternatively, the derivation means 123 uses, for example, the A* algorithm to calculate the starting point and the selected candidate destination so that the sum of the value of the repulsive force field and the value of the length of the path obtained as a result is minimized. At least one point between can be identified. For example, the further away from an obstacle the smaller the repulsive force field, but the longer the resulting path length. The A* algorithm can search for a point where the sum of the repulsive force field value w _p and the resulting path length value h is the minimum.

A first modified route is obtained by connecting the first starting point, the first selected candidate destination, and at least one point.

In process 900, steps S911 to S913 may be performed before step S901. Steps S911 to S912 are processes for deriving a second modified path for the second straight path. This makes it possible to obtain a first modified path by regarding the second modified path already obtained as an obstacle. The first modified path obtained in this manner can be a path that not only minimizes the possibility of collision with an actual obstacle that may exist between the first starting point and the first selection candidate destination point, but also minimizes the possibility of collision with the paths of other moving bodies.

In step S911, the deriving means 123 of the processor unit 120 calculates the repulsive force field that the obstacle exerts on the second straight path. Step S911 is the same process as step S901.

In step S912, the deriving means 123 of the processor unit 120 connects the second starting point, the second selected candidate arrival point, and at least one point based on the repulsive force field, thereby determining the second modified route. Derive. Step S912 is the same process as step S902.

In step S913, the deriving means 123 of the processor unit 120 regards the second corrected route derived in step S912 as an obstacle. For example, the second modified route may be expressed as a set of a plurality of obstacle points and stored in the memory unit 130 or the like.

Thereafter, in step S901, steps S901 and S902 are performed using the second modified route as an obstacle.

In the example shown in FIG. 9, it has been explained that in steps S911 to S913, a second corrected route is obtained for the second straight route and it is regarded as an obstacle, but in addition to the second straight route, , each corrected path may be obtained for at least one other straight path, and each corrected path may be considered as an obstacle.

FIG. 10 is a flowchart illustrating an example of a flow for planning travel routes for a plurality of moving bodies using the system 100 for creating travel routes for each of a plurality of mobile bodies.

In step S1001, planning of a travel route is started.

In step S1002, movement conditions are input into the system 100. For example, a user can input travel conditions into the system 200 via a terminal device. The movement conditions include, for example, a starting point and a plurality of candidate destination points for each of the plurality of moving bodies.

Thereby, the system 100 acquires the movement conditions. This corresponds to step S701 of process 700.

In step S1003, map information is acquired. For example, the system 100 acquires map information stored in the database unit 200 based on the travel conditions. The map information may include information on the terrain and buildings around the starting point and the destination point included in the travel conditions.

In step S1004, a three-dimensional moving space map is generated. For example, system 100 can generate a mobile spatial map based on map information.

In step S1005, movement prohibited area information is acquired. For example, the system 100 obtains no-movement zone information stored in the database unit 200 based on map information or a moving space map.

In step S1006, combinations of starting points and destination points of each of the plurality of moving bodies are determined on the two-dimensional plane. For example, by performing the process in step S702 of the process 700, the system 100 can determine the combination of the starting points and destination points of each of the plurality of moving objects, and the straight path connecting the starting points and the destination points. can.

In step S1007, a straight route is generated on the three-dimensional map. For example, the system 100 can generate a straight line route on a three-dimensional map by expanding the straight line route determined in step S1006 into a three-dimensional space.

In step S1008, a group of points to be placed around the straight path is calculated. For example, the system 100 can calculate the point group to be placed by performing steps S801 and S802 of the process 800. The point group to be placed may have, for example, a random point distribution, or a point distribution where the points become denser as they get closer to the center and become sparser as they move away from the center.

In step S1009, a point cloud of the obstacles and the generated path is generated to represent the obstacles and the generated path with multiple points. For example, the system 100 can use a KD tree to generate the point cloud of the obstacles and the generated path.

In step S1010, a moving object (and a corresponding starting point) whose route is to be determined is randomly selected from among moving objects whose route has not yet been determined. For example, system 100 randomly selects one mobile (and a corresponding starting point) among the mobiles whose route has not yet been determined. This corresponds, for example, to the situation shown in FIG. 6A.

In step S1011, the selected straight path of the moving object is excluded from the obstacle point group. For example, the system 100 excludes the selected moving object's straight path from the obstacle point cloud.

In step S1012, a point group is arranged around the starting point and arrival point of the moving body. for example. The system 100 can arrange the point cloud by performing steps S803 to S804 of the process 800. This corresponds, for example, to the situation shown in FIG. 6B.

In step S1013, a weighted minimum path search using A* is performed. For example, the system 100 performs step S703 of process 700 or steps S901 to S902 of process 900 to derive a path that minimizes the repulsive force field using the A* algorithm. This corresponds, for example, to the situation shown in FIGS. 6C and 6D.

In the weighted minimum path search using A*, the path connecting the starting point and the destination point may be searched at once, or may be searched in multiple stages. For example, the first subpath is identified by performing a weighted minimum path search using A* for the path from the starting point to a predetermined distance (also called search radius), and the first subpath is identified for the path from the end point of the first subpath to the predetermined distance. The second subpath is identified by performing a weighted minimum path search using A*, and this process can be repeated until the destination point is reached. The search radius may be, for example, 5%, 10%, 20%, 25%, etc. of the distance between the starting point and the destination point.

In step S1014, it is determined whether route generation has been completed for all moving objects. For example, the system 100 determines whether route generation has been completed for all moving objects.

If it is determined that route generation has been completed for all moving objects, the process advances to step S1015.

If it is determined that route generation has not been completed for all moving objects, the process returns to step S1009 and steps S1009 to S1014 are repeated. This is repeated until route generation is completed for all moving objects. This corresponds, for example, to the situation shown in FIG. 6E.

In step S1015, planning of the travel route ends.

In this way, routes for each of the plurality of moving objects are created. By moving each of the plurality of moving bodies according to the created route, the possibility of colliding with an obstacle and each of the other moving bodies is minimized.

In the example described above, each step of processing 700, 800, 900, and 1000 is performed in a specific order, but the described order is only an example. Each step of processes 700, 800, 900, 1000 can be performed in any logically possible order.

In the example described above with reference to FIGS. 7, 8, and 9, the processing of each step shown in FIGS. 7, 8, and 9 is realized by the processor unit 120 and the program stored in the memory unit 130. However, the present invention is not limited thereto. At least one of the processes in each step shown in FIGS. 7, 8, and 9 may be realized by a hardware configuration such as a control circuit.

The present invention is not limited to the embodiments described above. It is understood that the invention is to be construed in scope only by the claims. It will be understood that those skilled in the art will be able to implement the present invention to an equivalent extent based on the description of the present invention and common general technical knowledge from the description of the specific preferred embodiments of the present invention.

An application was developed to implement the process 700 for creating travel routes for each of a plurality of moving objects. FIG. 11 shows an example of a user interface 1100 of an application for implementing the process 700 for creating travel routes for each of a plurality of mobile objects. Here, an example will be explained in which a movement route is created for each of a plurality of flying objects (here, unmanned aerial vehicles (UAVs)).

When the user inputs conditions for creating travel routes for each of the plurality of flying objects into the user interface 1100, this application can create travel routes and display them to the user.

The user interface 1100 includes a position data input button (Position Data) 1101, an obstacle data input button (Obstacle Data) 1102, a drone setting button (Drones Ctrl.) 1103, an execute button 1104, a save button 1105, a path parameter input section (Path Calculation) 1106, a flight parameter input section (Flight Parameters) 1107, and an initial position/arrival position input section (Drone Points) 1108. The map displayed on the user interface 1100 shows the navigation center position 1109 and the navigation range 1110.

FIG. 12 is a flowchart illustrating an example of an application operating method for implementing processing 700 for creating movement routes for each of a plurality of moving objects.

First, in step S1201, the user sets the navigation center position. The user can set the navigation center position by moving the navigation center position 1109 displayed on the user interface 1101. This allows the user to intuitively set the navigation center position.

Next, in step S1202, the user sets the navigation range. The user can set the navigation range by moving and/or enlarging/reducing the navigation range 1110 displayed on the user interface 1101. This allows the user to intuitively set the navigation range.

At this time, the user can set route parameters and flight parameters via a path parameter input section (Path Calculation) 1106 and a flight parameter input section (Flight Parameters) 1107. Route parameters and flight parameters are part of the movement conditions mentioned above.

Furthermore, at this time, the user may input obstacle data via an obstacle data input button (Obstavle Data) 1102. Thereby, obstacles can be set within the navigation range. The obstacle data is, for example, data on a plurality of obstacle points.

Next, in step S1203, the user sets the number of UAVs to be flown. The user can increase or decrease the number of UAVs to be flown by operating the [+] or [-] of the flying object setting button (Drones Ctrl.) 1103.

Next, in step S1204, the user sets the initial position and destination position of the UAV to be flown. The user can input the initial position and destination position in [x] and [y] of the initial position/destination position input section (Drone Points) 1108. The user can, for example, input text in [x] and [y] of the initial position/destination position input section (Drone Points) 1108. Alternatively, the user can input data representing the initial position and destination position of the UAV via the position data input button (Position Data) 1101.

Next, in step S1205, the user sets the takeoff altitude and arrival altitude of the UAV to be flown. The user can input the takeoff altitude and the arrival altitude into [Alt] of the initial position/arrival position input section (Drone Points) 1108. The user can input text to [Alt] of the initial position/arrival position input section (Drone Points) 1108, for example. Alternatively, the user can enter data representative of the UAV's takeoff altitude and arrival altitude via a Position Data input button (Position Data) 1101.

Next, in step S1206, the user sets a takeoff time difference for the UAV to be flown. The user can input the takeoff altitude and arrival altitude into [Delay] of the initial position/arrival position input section (Drone Points) 1108. The user can input text to [Delay] of the initial position/arrival position input section (Drone Points) 1108, for example.

Next, in step S1207, the user reflects each setting of the UAV to be flown. The user confirms the settings input so far by operating the [!] of the drone setting button (Drones Ctrl.) 1103. This causes the application to receive the movement conditions of the UAV.
For example, if the user wishes to finely adjust the navigation center position, in step S1208, the user can reset the navigation center position by performing the same operation as in step S1201. In this case, in step S1207,

You need to go back and reflect the reconfigured settings.

Next, in step S1209, the user activates the automatic path construction function. The user can activate the automatic path construction function by operating the execute button 1104. This causes process 700 to be performed within the application.

Upon completion of process 700, the created flight path is presented to the user. The flight path may be presented to the user in a 3D and/or 2D display.

FIG. 13 shows an example of a screen 1300 that includes a 3D display of a flight path presented to a user.

Screen 1300 displays the four created flight paths 1310 on a 3D rendered topographical map.

Referring again to FIG. 12, in step S1210, the user visually checks the created flight route on the 3D display to check whether there are any intersections or the like. For example, visually check the screen 1300 in FIG. 13 to confirm that there is no intersection.

In step S1211, the user visually checks the created flight route on the 2D display and checks whether there are any intersections or the like. This is because even if there is no intersection on the 3D display, there is a possibility that there is an intersection on the 2D display. For example, if they do not intersect in three-dimensional space but do intersect on a two-dimensional plane, there is a possibility that the aircraft will collide with each other if a sudden descent or steep rise occurs due to the occurrence of downdraft/updraft or the occurrence of an aircraft abnormality. Therefore, it is important to confirm that they do not intersect even on a 2D display. As described above, in the process 700, a flight path can be created such that the number of intersections on a two-dimensional plane is minimized, so it is unlikely that a flight path that intersects on a 2D display will be created.

If there is a defect such as an intersection in the created flight route, the process can return to step S1204 and repeat steps S1204 to S1211.

If the created flight path is sufficient, in step S1212, the user can output the created flight path. The user can save the created flight route by operating the save button 1105. The flight route can be output as CSV data, for example.

INDUSTRIAL APPLICABILITY The present invention is useful in providing a method, system, and program for creating movement routes for a plurality of moving bodies, which minimize collisions between the plurality of moving bodies.

U User S Departure G Destination 10 Aircraft 100 System 110 Interface section 120 Processor section 130 Memory section 200 Database section

Claims (14)

A method of creating a movement route for each of a plurality of moving bodies including at least a first moving body and a second moving body, the creation method being executed in a computer system having a processor unit, the creation method teeth,
(1) The processor unit acquires movement conditions including a starting point and a plurality of candidate destination points for each of the plurality of moving objects, and the starting point is for the first moving object. a first starting point and a second starting point for the second mobile object , and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile object and a second starting point for the second mobile object. a second plurality of candidate arrival points for the second mobile body ;
(2) the processor section determines a straight path for each of the plurality of moving objects,
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second mobile object by connecting the second starting point and a second selected candidate destination paired with the second starting point; The above-mentioned pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another starting point of the starting point group. A depth-first search method with pruning is used to count the number of pairwise intersections so that the number of intersections between the starting point and another candidate arrival point of the end point group is minimized. searching for a pair of each starting point of a group and any one of the ending point groups;
(3) The processor unit, for at least the first linear route, determines the first starting point, the first selected candidate destination point, and between the starting point and the first selected candidate destination point. deriving a first modified path by connecting at least one point;
(2) Determining the straight path is as follows:
identifying a straight line that does not intersect with any other straight line path;
A point group including the first starting point, the second starting point, and a candidate arrival point group including the first plurality of candidate arrival points and the second plurality of candidate arrival points is dividing by a straight line,
Identifying the first selection candidate destination point within the point group that includes the first starting point among the point group after division;
specifying the second selection candidate arrival point in a point group that includes a second starting point among the point group after division, and the one straight line A creation method in which the number of starting points and the number of candidate destination points are specified to be the same in at least one of the above . The creation method according to claim 1 , wherein the intersection is an intersection on a two-dimensional plane. The method according to claim 1 , wherein the at least one point is a point in a cloud of points arranged around the first linear path. A method of creating a movement route for each of a plurality of moving bodies including at least a first moving body and a second moving body, the creation method being executed in a computer system having a processor unit, the creation method teeth,
(1) The processor unit acquires movement conditions including a starting point and a plurality of candidate destination points for each of the plurality of moving objects, and the starting point is for the first moving object. a first starting point and a second starting point for the second mobile object , and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile object and a second starting point for the second mobile object. a second plurality of candidate arrival points for the second mobile body ;
(2) the processor section determines a straight path for each of the plurality of moving objects,
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second mobile object by connecting the second starting point and a second selected candidate destination paired with the second starting point; The above-mentioned pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another starting point of the starting point group. A depth-first search method with pruning is used to count the number of pairwise intersections so that the number of intersections between the starting point and another candidate arrival point of the end point group is minimized. searching for a pair of each starting point of a group and any one of the ending point groups;
(3) The processor unit, for at least the first linear route, determines the first starting point, the first selected candidate destination point, and between the starting point and the first selected candidate destination point. deriving a first modified path by connecting at least one point;
the at least one point is a point in a point cloud arranged around the first linear path;
The plurality of points in the point group are distributed such that the closer the points are to the first starting point, the denser the points are, and/or the closer the points are to the first selected candidate destination point, the denser the points are. Yes, how to create it. The point group is
Obtaining a point cloud with a random distribution in a polar coordinate system;
Converting a point group in the polar coordinate system to a point group in a rectangular coordinate system;
centering the transformed point cloud on the first starting point;
The creation method according to claim 3 or 4 , wherein the transformed point group is arranged by centering the first selection candidate arrival point. A method of creating a movement route for each of a plurality of moving bodies including at least a first moving body and a second moving body, the creation method being executed in a computer system having a processor unit, the creation method teeth,
(1) The processor unit acquires movement conditions including a starting point and a plurality of candidate destination points for each of the plurality of moving objects, and the starting point is for the first moving object. a first starting point and a second starting point for the second moving object , and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first moving object and a second starting point for the second moving object. a second plurality of candidate arrival points for the second mobile body ;
(2) the processor section determines a straight path for each of the plurality of moving objects,
The starting points of each of the plurality of moving bodies are defined as a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are defined as a final point group, and each starting point of the plurality of starting points and which one of the terminal point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination point paired with the first starting point;
determining a second straight path of the second mobile object by connecting the second starting point and a second selected candidate destination paired with the second starting point; The pairing means that a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another starting point of the starting point group. A depth-first search method with pruning is used to count the number of pairwise intersections so that the number of intersections between the starting point and another candidate arrival point of the end point group is minimized. searching for a pair of each starting point of the group and any one of the ending point group;
(3) The processor unit, for at least the first linear route, determines the first starting point, the first selected candidate destination point, and the distance between the starting point and the first selected candidate destination point. deriving a first modified path by connecting at least one point;
Deriving the first modified path includes:
(i) calculating a repulsive force field exerted by an obstacle on the first straight path;
(ii) Connecting the first starting point, the first selection candidate arrival point, and the at least one point so that the repulsive force field is minimized. Deriving the first modified path includes:
(a) calculating a repulsive force field exerted by an obstacle on the second straight path;
(b) based on the repulsive force field, at least one of the second starting point, the second selected candidate destination point, and the second selected candidate destination point between the second starting point and the second selected candidate destination point; deriving a second modified path by connecting the points;
(c) deriving the first modified route by performing (i) and (ii) on the first straight route using the second modified route as an obstacle; , The production method according to claim 6 . The creation method according to any one of claims 1 to 7 , wherein the first moving object and the second moving object are randomly selected from the plurality of moving objects. A system for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the system comprising:
Acquisition means for acquiring movement conditions including a starting point and a plurality of candidate arrival points for each of a plurality of moving objects, wherein the starting point is a first starting point for the first moving object and a plurality of candidate arrival points. a second starting point for the second mobile body, and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile body and a second starting point for the second mobile body. 2. a plurality of candidate arrival points ;
A determining means for determining a straight line route for each of the plurality of moving objects, the determining means comprising:
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second moving body by connecting the second starting point and a second selected candidate arrival point paired with the second starting point; The pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another one of the starting point group. A depth-first search method with pruning is used to count the number of pairwise intersections so that the number of intersections with a straight line connecting one starting point and another candidate arrival point of the group of destination points is minimized. , searching for a pair of each starting point of the starting points and any one of the ending points ;
connecting the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point for at least a first straight path; derivation means for deriving the first modified path by; the derivation means comprising:
(i) calculating a repulsive force field exerted by an obstacle on the first straight path;
(ii) connecting the first starting point, the first selected candidate destination, and the at least one point such that the repulsive force field is minimized. A program for generating a moving path for each of a plurality of moving bodies including at least a first moving body and a second moving body, the program being executed in a computer system having a processor unit, the program comprising:
(1) acquiring travel conditions including a starting point and a plurality of candidate destination points for each of a plurality of moving bodies, the starting point including a first starting point for the first moving body and a second starting point for the second moving body, and the plurality of candidate destination points including a first plurality of candidate destination points for the first moving body and a second plurality of candidate destination points for the second moving body;
(2) determining a straight line path for each of the plurality of moving objects,
A starting point of each of the plurality of moving bodies is defined as a starting point group, a plurality of candidate arrival points of each of the plurality of moving bodies is defined as an end point group, and each starting point of the starting point group is paired with any one of the candidate arrival points of the end point group;
determining a first straight-line path for a first moving object by connecting the first starting point and a first selected candidate destination point paired with the first starting point;
determining a second straight-line path of a second moving object by connecting the second starting point with a second selected candidate destination point paired with the second starting point , the pairing including searching for pairs of each starting point in the group of starting points and any one of the group of destination points using a pruning depth-first search method that counts the number of pair intersections between a straight line connecting one starting point in the group of starting points and one candidate destination point in the group of destination points and a straight line connecting another starting point in the group of starting points and another candidate destination point in the group of destination points, so as to minimize the number of intersections between the straight line connecting one starting point in the group of starting points and one candidate destination point in the group of destination points ;
(3) for at least a first straight line path, deriving a first modified path by connecting the first starting point, the first selection candidate destination point, and at least one point between the starting point and the first selection candidate destination point, and deriving the first modified path includes:
(i) calculating a repulsive force field that an obstacle exerts on the first linear path;
(ii) connecting the first starting point, the first selection candidate destination point, and the at least one point so that the repulsive field is minimized. A system for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the system comprising:
Acquisition means for acquiring movement conditions including a starting point and a plurality of candidate arrival points for each of a plurality of moving objects, wherein the starting point is a first starting point for the first moving object and a plurality of candidate arrival points. a second starting point for the second mobile body, and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile body and a second starting point for the second mobile body. 2. a plurality of candidate arrival points;
A determining means for determining a straight line route for each of the plurality of moving objects, the determining means comprising:
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second moving body by connecting the second starting point and a second selected candidate destination paired with the second starting point;
The pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and the other of the starting point group. A pruned depth-first search method is used to count the number of pairwise intersections so that the number of intersections with a straight line connecting one starting point and one other candidate arrival point of the end point group is minimized. determining means for searching for a pair of each starting point of the group of starting points and any one of the group of ending points using the method;
connecting the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point for at least a first straight path; a deriving means for deriving the first modified path by
including;
The determining means is
identifying a straight line that does not intersect with any other straight line path;
A point group including the first starting point, the second starting point, and a candidate arrival point group including the first plurality of candidate arrival points and the second plurality of candidate arrival points is dividing by a straight line,
Identifying the first selection candidate destination point within the point group that includes the first starting point among the point group after division;
specifying the second selection candidate destination point within the point group that includes the second starting point among the point group after the division;
and the one straight line is specified such that the number of starting points and the number of candidate destination points are the same in at least one of the point groups after division. A program for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the program being executed in a computer system having a processor section, The program is
(1) Obtaining movement conditions including a starting point and a plurality of candidate arrival points for each of a plurality of moving objects, wherein the starting point is a first starting point for the first moving object. and a second starting point for the second mobile body, and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile body and a second starting point for the second mobile body. a second plurality of candidate destinations; and
(2) determining a straight path for each of the plurality of moving objects,
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second moving body by connecting the second starting point and a second selected candidate destination paired with the second starting point;
and the pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another starting point of the starting point group. Using a depth-first search method with pruning that counts the number of pairwise intersections so that the number of intersections between a point and a straight line connecting another candidate arrival point of the end point group is minimized, searching for a pair of each starting point of the starting points and any one of the ending points;
(3) for at least the first straight route, the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point; Derive the first modified path by connecting
causing the processor unit to perform processing including;
(2) Determining the straight path is as follows:
identifying a straight line that does not intersect with any other straight line path;
A point group including the first starting point, the second starting point, and a candidate arrival point group including the first plurality of candidate arrival points and the second plurality of candidate arrival points is dividing by a straight line,
Identifying the first selection candidate destination point within the point group that includes the first starting point among the point group after division;
specifying the second selection candidate destination point within the point group that includes the second starting point among the point group after the division;
, wherein the one straight line is specified such that the number of starting points and the number of candidate destination points are the same in at least one of the point groups after division. A system for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the system comprising:
Acquisition means for acquiring movement conditions including a starting point and a plurality of candidate arrival points for each of a plurality of moving objects, wherein the starting point is a first starting point for the first moving object and a plurality of candidate arrival points. a second starting point for the second mobile body, and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile body and a second starting point for the second mobile body. 2. a plurality of candidate arrival points;
A determining means for determining a straight line route for each of the plurality of moving objects, the determining means comprising:
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second moving body by connecting the second starting point and a second selected candidate destination paired with the second starting point;
The pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and the other of the starting point group. A pruned depth-first search method is used to count the number of pairwise intersections so that the number of intersections with a straight line connecting one starting point and one other candidate arrival point of the end point group is minimized. determining means for searching for a pair of each starting point of the group of starting points and any one of the group of ending points using the method;
connecting the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point for at least a first straight path; a deriving means for deriving the first modified path by
including;
the at least one point is a point in a point cloud arranged around the first linear path;
The plurality of points in the point group are distributed such that the closer the points are to the first starting point, the denser the points are, and/or the closer the points are to the first selected candidate destination point, the denser the points are. There is a system . A program for creating movement routes for each of a plurality of moving bodies including at least a first moving body and a second moving body, the program being executed in a computer system having a processor section, The program is
(1) Obtaining movement conditions including a starting point and a plurality of candidate arrival points for each of a plurality of moving objects, wherein the starting point is a first starting point for the first moving object. and a second starting point for the second mobile body, and the plurality of candidate arrival points include a first plurality of candidate arrival points for the first mobile body and a second starting point for the second mobile body. a second plurality of candidate destination points; and
(2) determining a straight path for each of the plurality of moving objects,
The starting point of each of the plurality of moving bodies is a starting point group, and the plurality of candidate arrival points of each of the plurality of moving bodies are a destination point group, and each starting point of the starting point group and which one of the ending point group or one candidate destination, and
determining a first straight path of the first mobile object by connecting the first starting point and a first selected candidate destination paired with the first starting point;
determining a second straight path of the second moving body by connecting the second starting point and a second selected candidate destination paired with the second starting point;
and the pairing includes a straight line connecting one starting point of the starting point group and one candidate arrival point of the ending point group, and another starting point of the starting point group. Using a depth-first search method with pruning that counts the number of pairwise intersections so that the number of intersections between a point and a straight line connecting another candidate arrival point of the end point group is minimized, searching for a pair of each starting point of the starting points and any one of the ending points;
(3) for at least the first straight route, the first starting point, the first selected candidate destination point, and at least one point between the starting point and the first selected candidate destination point; Derive the first modified path by connecting
causing the processor unit to perform processing including;
the at least one point is a point in a point cloud arranged around the first linear path;
The plurality of points in the point group are distributed such that the closer the points are to the first starting point, the denser the points are, and/or the closer the points are to the first selected candidate destination point, the denser the points are. Yes, there is a program .