3D transport path search device and 3D transport path search method

Abstract

To provide a 3D transportation route search device capable of performing route search with high accuracy.SOLUTION: A 3D transportation route search device includes: a database 102 for storing extended transportation route information 207 on an extended movement route obtained by giving a first width based on a position error of a moving object to a transportation route of the moving object and information on a 3D map 204 including information on a size of an obstacle; and a transportation route search unit for generating a 3D transportation route in which the moving object and a transportation object do not collide with the obstacle based on the extended transportation route information 207 and the information of the 3D map 204 stored in the database 102.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a three-dimensional conveyance route search device and a three-dimensional conveyance route search method.

Background Art

[0002] As devices for conveying a conveyed object (hereinafter referred to as a moving object), there are various devices such as drones and ceiling cranes. When moving a moving object in a three-dimensional space or conveying a conveyed object with a moving object, it may be necessary to confirm in advance what the conveyance route will be or to determine in advance the control in the three-dimensional direction by feedforward control or the like. In that case, before starting the conveyance, a three-dimensional conveyance route is specified, and it is confirmed in advance that the moving object and the conveyed object do not interfere with obstacles on the route.

[0003] For example, in Patent Document 1, a ceiling crane is described as an example of a moving object, and a predetermined margin space is set for the conveyed object or an object on the conveyance route in order to avoid a collision between the conveyed object and an obstacle when moving on the route.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] By the way, when specifying the three-dimensional conveyance route, it is necessary to specify the three-dimensional conveyance route in consideration of the following errors. For example, there are numerical errors in the conveyance route obtained by numerical calculation. In addition to errors due to control deviations caused by noise or the like in actual conveyance, errors due to deviations between the route such as the theoretically shortest conveyance route and the actual conveyance, the swing width of the conveyed object on the traveling route, and errors in the height and position information of obstacles on the route where the moving object can move.

[0006] Patent Document 1, mentioned above, describes how to estimate the amplitude of the movement of a conveyed object and set a safety margin. However, this safety margin does not take into account the errors in the moving object as described above, and it is not possible to accurately confirm the possibility of collision between the conveyed object and an obstacle. [Means for solving the problem]

[0007] A three-dimensional transport path search device according to an aspect of the present invention searches for a three-dimensional transport path of a moving object that transports an object. thing And, Information about the moving object, including its speed, acceleration and deceleration, the mass of the moving object, and the height of the moving object from the ground at the start of transport; and information about the upward and downward control patterns, which may or may not describe a plurality of upward and downward control patterns representing the methods of raising and lowering the transported object. This is obtained by adding a first width based on the position error of the moving object to the movement path of the moving object. on a 2D plane expansion 2D Travel route Extended travel path information representing, Obstacle Location and 3D map information including size information and, A storage unit that stores and the storage unit The moving object information, the upward / downward control pattern information, The extended travel path information The system also includes a transport path search unit that generates a three-dimensional transport path in which the moving object and the transported object do not collide with obstacles, based on the three-dimensional map information. If the ascending / descending control pattern information contains the multiple ascending / descending control patterns, the transport path search unit selects an ascending / descending control pattern from the multiple ascending / descending control patterns represented by the ascending / descending control pattern information that, when the transport of the moving object is carried along the extended two-dimensional movement path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional movement path, and minimizes the overall transport time, thereby generating the three-dimensional transport path. If the ascending / descending control pattern information does not contain the multiple ascending / descending control patterns, the transport path search unit searches for an ascending / descending control pattern based on the moving object information, the extended movement path information, and the three-dimensional map information that, when the transport of the moving object is carried along the extended two-dimensional movement path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional movement path, and minimizes the overall transport time, thereby generating the three-dimensional transport path. . A three-dimensional transport path search method according to an aspect of the present invention is: A method for searching for a three-dimensional transport path of a moving object that transports an object, using a computer and a database, the In the database This includes moving object information including information on each direction in which the moving object can move, the maximum speed during acceleration and deceleration for each direction of the moving object, acceleration and deceleration, the mass of the moving object, and the height of the moving object from the ground at the start of transport, and upward / downward control pattern information which may or may not describe a plurality of upward / downward control patterns that represent the method of raising and lowering the transported object, and the Obtained by adding a width to the movement path of the moving object based on the position error of the moving object. on a 2D plane expansion 2D Travel route Extended travel path information representing, Obstacle Location and 3D map information including size information If the above and below information is stored and the above and below control pattern information contains the above and below control patterns, the computer selects from the above and below control patterns represented by the above and below control pattern information that, when the transport of the moving object along the extended two-dimensional movement path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional movement path and minimizes the overall transport time, thereby generating the three-dimensional transport path. If the above and below control pattern information does not contain the above and below control patterns, the computer searches for an above and below control pattern based on the moving object information, the extended movement path information, and the three-dimensional map information that, when the transport of the moving object along the extended two-dimensional movement path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional movement path and minimizes the overall transport time, thereby generating the three-dimensional transport path. . [Effects of the Invention]

[0008] According to the present invention, by considering errors related to the moving object when searching for a 3D transport path that can avoid collisions, it is possible to perform a highly accurate 3D transport path search. [Brief explanation of the drawing]

[0009] [Figure 1]FIG. 1 is a block diagram showing an example of the functional configuration of a three-dimensional conveyance path search device. [Figure 2] FIG. 2 is a diagram showing an example of the hardware configuration of a three-dimensional conveyance path search device. [Figure 3] FIG. 3 is a diagram showing a list of parameters stored in a database. [Figure 4] FIG. 4 is a diagram showing a specific example of moving object information. [Figure 5] FIG. 5 is a diagram showing a specific example of error information. [Figure 6] FIG. 6 is a diagram showing a specific example of conveyance path information. [Figure 7] FIG. 7 is a diagram showing a specific example of a three-dimensional processing width and a path width. [Figure 8] FIG. 8 is a diagram showing a specific example of extended conveyance path information. [Figure 9] FIG. 9 is a diagram showing a specific example of three-dimensional conveyance information. [Figure 10] FIG. 10 is a diagram showing a specific example of a three-dimensional map. [Figure 11] FIG. 11 is a diagram showing a specific example of a processed three-dimensional map. [Figure 12] FIG. 12 is a diagram showing a specific example of three-dimensional conveyance distance information. [Figure 13] FIG. 13 is a diagram showing specific examples of a path creation method and an up / down control pattern. [Figure 14] FIG. 14 is a flowchart showing a processing procedure in a path search unit. [Figure 15] FIG. 15 is a flowchart showing detailed processing of step S803. [Figure 16] FIG. 16 is a flowchart showing detailed processing of step S804. [Figure 17] FIG. 17 is a flowchart showing detailed processing of step S805. [Figure 18] FIG. 18 is a flowchart showing detailed processing of step S806. [Figure 19]Figure 19 is a flowchart showing the detailed processing of step S807. [Figure 20] Figure 20 is a flowchart showing the detailed processing of step S808. [Figure 21] Figure 21 is a schematic diagram showing the overhead crane transport in Example 1. [Figure 22] Figure 22 is a schematic diagram showing a two-dimensional path drawn on a two-dimensional map. [Figure 23] Figure 23 shows the three-dimensional transport distance information in Example 1. [Figure 24] Figure 24 shows an example of the output of a path in a two-dimensional plane in Example 1. [Figure 25] Figure 25 shows the extended transport path from Figure 24, with the horizontal axis representing the direction of movement of the moving object and the vertical axis representing its height. [Figure 26] Figure 26 is a schematic diagram showing the general outline of drone-based transport in Example 2. [Figure 27] Figure 27 shows a two-dimensional map obtained by projecting the three-dimensional map from Figure 26 onto the xy-plane. [Figure 28] Figure 28 shows the three-dimensional transport distance information in Example 2. [Figure 29] Figure 29 shows the 3D processing map and transport route in Example 2. [Modes for carrying out the invention]

[0010] The embodiments for carrying out the present invention will be described below with reference to the figures. The following description and drawings are illustrative examples for explaining the present invention, and have been omitted and simplified as appropriate for clarity of explanation. In addition, in the following description, the same or similar elements and processes are denoted by the same reference numerals, and redundant explanations may be omitted. It should be noted that the contents described below are merely examples of embodiments of the present invention, and the present invention is not limited to the embodiments described below, and can be carried out in various other forms.

[0011] <Functional Configuration of a 3D Transport Pathfinding System> Figure 1 is a block diagram showing an example of the functional configuration of a 3D transport path search device. Hereafter, the 3D transport path search device will be simply referred to as a path search device. The path search device 100 is a device that searches for a 3D transport path based on a specific path of a moving object. In the example shown in Figure 1, the path search device 100 has a data input / output unit 101, a database 102, and a path search unit 103.

[0012] The route search device 100 can access the database 102. The database 102 is stored in a computer within the route search device 100 or in an external computer that can communicate with the route search device 100. The database 102 stores the parameters shown in Figure 3, which will be described later. The database 102 also stores various information calculated by the route search unit 103.

[0013] The data input / output unit 101 accepts data input from an input device such as a keyboard, or from an external computer that is connected to the route search device 100 in a communicative manner. The data input / output unit 101 also outputs data calculated within the route search device 100 to an output device such as a display, or to an external computer in a displayable format. The route search unit 103 includes, as processing units, a route creation unit 110, an error estimation unit 111, a 3D map processing unit 112, an extended route creation unit 113, a transport route search unit 114, and an output result creation unit 115.

[0014] <Hardware configuration of the route search device 100> Figure 2 shows an example of the hardware configuration of the route search device 100 shown in Figure 1. The route search device 100 comprises a processor 701, a data storage device 702, a communication device 703, an input device 704, an output device 705, and a program storage device 706 as its hardware configuration.

[0015] The processor 701 is a processor such as a CPU (Central Processing Unit) that controls the pathfinding device 100. The data storage device 702 is the work area for the processor 701. The data storage device 702 and the program storage device 706 are non-temporary or temporary recording media that store data and various programs. Examples of data storage devices 702 include ROM (Read Only Memory), RAM (Random Access Memory), HDD (Hard Disk Drive), and flash memory. The program storage device 706 stores programs for executing the processing of each part of the pathfinding unit 103.

[0016] The communication device 703 connects to the network and sends and receives data. The input device 704 is a device that inputs data, such as a keyboard, mouse, touch panel, numeric keypad, scanner, microphone, etc. The output device 705 is a device that outputs data, such as a display, printer, speaker, etc.

[0017] The pathfinding unit 103 shown in Figure 1 is implemented by having the processor 701 execute programs stored in the data storage device 702 and the program storage device 706. The database 102 is stored in the data storage device 702. Note that an external computer's storage device may be used as the data storage device 702.

[0018] <Detailed description of database 102> Figure 3 shows a list of parameters stored in database 102. Figures 4 to 12 show detailed examples of the parameters in Figure 3. Database 102 includes moving object information 200, error information 201, transport route information 202, 3D map processing width 203, 3D map 204, processed 3D map 205, route width 206, extended transport route information 207, 3D transport information 208, 3D transport distance information 209, 3D transport route 210, route creation method 211, and ascent / descent control pattern 212.

[0019] (Moving object information 200) The moving object information 200 includes information such as the maximum speed, acceleration, and deceleration of the moving object in each direction, as well as its height from the ground and mass at the start of transport. The moving object is a device used to transport transported objects, and as described later, this includes overhead crane trolleys and drones that fly in three-dimensional space. The moving object information 200 also includes information such as the size and mass of the transported object in each direction and its height from the ground at the start of transport. In the case of a moving object such as a drone, the transported object is directly held by the drone and exhibits the same movement as the moving object, so in such cases, a flag indicating that the transported object and the moving object are the same is stored in the moving object information 200.

[0020] (Specific example of moving object information 200) Figure 4 shows a specific example 301 of the moving object information 200. The moving object information 200 includes the ID of the moving object, the directions in which the moving object can move, the maximum speed during acceleration and deceleration in each direction, the acceleration and deceleration during acceleration and deceleration, the height of the moving object from the ground, the mass of the moving object, the size and mass of the transported object, and flag information indicating whether the transported object and the moving object are the same.

[0021] In the example described in Specific Example 301, the ID of the moving object, the height of the moving object from the ground, the mass of the moving object, the mass of the transported object, and the flag information indicating whether the transported object and the moving object are the same object are independent of the direction of movement of the moving object. However, values ​​for other parameters such as maximum speed (acceleration), maximum speed (deceleration), acceleration, deceleration, and the size of the transported object are stored for each direction. The size of the transported object in each direction may be the longest distance in that direction, the average value, the minimum value, or any other value.

[0022] If there are multiple moving objects in an area where a particular moving object can move, each moving object will be assigned a different ID. Regarding the flag used to determine whether a transported object and a moving object are the same, "True" indicates they are the same, while "False" indicates they are different. If the size and mass information of a transported object is 0 (for example, the case where ID=1 in Figure 4), even if the flag determining whether the transported object and the moving object are the same is "True," the moving object is considered to have no transported object and will not be involved in transport during pathfinding. Furthermore, for moving objects that are installed, such as overhead cranes, the height from the ground should be recorded. For objects whose height information changes, such as drones, the height at the start of transport or a value greater than or equal to 0 should be recorded. Note that in the example 301, the units of each piece of information are displayed in [mm] and [kg], but these are just examples and the system is not limited to these units.

[0023] (Error information 201) Error information 201 includes numerical errors (fixed, time) when a specific path in a two-dimensional plane (xy-plane) is numerically determined, numerical errors during pathfinding in three dimensions, errors between transport passing through a specific path and actual transport, observation errors from sensors, etc., when creating a three-dimensional map, and information that can estimate information errors when the information on the three-dimensional map may change after the three-dimensional map is created. The information that can estimate information errors includes errors at each time that may occur in numerical calculations, and information that takes into account the difference between transport passing through a specific path and actually possible transport (i.e., the difference in transport time, speed, and position).

[0024] (Specific example of error information 201) Figure 5 shows a specific example 302 of error information 201. Error information 201 includes, for each ID of a moving object, time interval, 2D path numerical error (fixed), 2D path numerical error (time), 3D direction numerical error (fixed), 3D direction numerical error (time), difference from actual transport, 3D map sensor error (x, y, height), 3D map information error, etc.

[0025] 2D path numerical error (fixed) is a time-independent numerical error relating to the position of a path when a specific path in a 2D plane is numerically defined. On the other hand, 2D path numerical error (time) is a numerical error that changes at regular intervals, such as an accumulation error. 3D directional numerical error (fixed) and 3D directional numerical error (time) are errors in the z-direction when a moving object moves in the z-direction. Similar to the 2D path numerical error, 3D directional numerical error (fixed) is a time-independent numerical error when transport in the 3D direction is numerically defined, while 3D directional numerical error (time) is a numerical error that changes at regular intervals. A numerical error (time) is added to the common error (fixed) at each time interval. The time interval can be 1 second, 0.1 seconds, 1 minute, etc.

[0026] Numerical error may be an error estimated using numerical methods such as the trapezoidal rule or Simpson's method, which are used in the numerical calculation of differential equations, or it may be expressed as the degree of error, for example, an error to two decimal places being 0.01. Furthermore, the numerical error (fixed) may be set to 0 and the numerical error (time) to non-zero, or vice versa. In addition, the numerical error (time) may be calculated not only by addition but also by other arithmetic operations such as multiplication, or it may be an error calculated by inputting the stated error into a function.

[0027] Regarding the difference in actual transport, for example, if a specific route is assumed to be transported using the theoretically shortest transport time, there may be a discrepancy between the actual transport time and that theoretically shortest transport time. In other words, the difference in actual transport is the difference between the transport assumed when determining a specific route and the actual transport. For example, if the difference in actual transport is expressed as a percentage, a delay is indicated by a positive [%], and a speed is indicated by a negative [%]. Of course, it does not have to be expressed as a percentage; for example, the difference in transport position can be expressed in [mm], or any value that expresses the difference from the actual transport is acceptable.

[0028] 3D map sensor error (x, y, height) refers to the observation error by the sensor during 3D map creation. 3D map information error refers to the error between the time of 3D map creation and the time of actual transport, as the obstacle situation (3D map information) may change over time. Note that if the x and y directions differ due to 2D path numerical error (fixed, time) or differences in actual transport, the errors for each direction may also be listed.

[0029] (Transportation route information 202) The transport route information 202 is route information specified by the user, and includes the transport start position and transport end position in three-dimensional space and route information connecting each of these positions (the position of the moving object in the x and y directions). If the user does not specify any route other than the transport start position and transport end position, the route creation unit 110 creates a route connecting the transport start position and the transport end position (a route in a two-dimensional plane).

[0030] (Specific example of transport route information 202) Figure 6 shows a specific example 303 of the transport route information 202. The transport route information 202 includes information such as the ID of the moving object, the position of the moving object (x direction) at each time point, the position of the moving object (y direction), the initial height of the transported object, a transport start flag, and a transport end flag. In the specific example 303 of Figure 6, the time intervals are 0.1 seconds, but these intervals do not have to be equal. ID=1 is an example where no route information other than the transport start position and transport end position is provided. The time at the end of transport is unknown, so it is represented as "-" with no information, but it may be represented by another notation such as "None" or "-1".

[0031] For the transport start and transport end flags, if the object's position at that time indicates the start of transport, the transport start flag is denoted as "True" and the transport end flag as "False." If the object's position at that time indicates the end of transport, the transport end flag is denoted as "True" and the transport start flag as "False." Otherwise, "False" is used. The units for position and height are not limited to [mm]; they may also be [m]. The flag notation may also be replaced with expressions such as 0 and 1 instead of "True" and "False."

[0032] (3D map processing width 203) The 3D map processing width 203 is an error width specified by the user separately from the width of the error information 201 when the 3D map processing unit 112 creates the processed 3D map. The specific example 401 shown in Figure 7 is a specific example of the 3D map processing width 203. Width (x), width (y), and height (z) are widths specified by the user, and error consideration (x), error consideration (y), and error consideration (z) are flag information that determines whether or not to consider the error information 201, as shown in specific example 302, for obstacles in the 3D map.

[0033] The error information 201 considered for obstacles in the 3D map includes 3D map sensor errors and 3D map information errors, as shown in specific example 302. Regarding the consideration of errors in the x, y, and z directions as shown in specific example 401, if "True" is selected, the 3D map processing unit 112 considers the errors in each direction of the 3D map estimated by the error estimation unit 111. On the other hand, if "False" is selected, the 3D map processing unit 112 does not consider the errors estimated by the error estimation unit 111 when creating the processed 3D map 205, as described later, even if the errors are finite.

[0034] (Path width 206) The path width 206 is an error width specified by the user separately from the error information 201 when the extended path creation unit 113 creates an extended two-dimensional path.

[0035] Specific example 402 shown in Figure 7 is an example of a path width of 206. Width (x), width (y), and height (z) are widths specified by the user, and error consideration (perpendicular to the direction of travel), error consideration (direction of travel), and error consideration (z) are flag information. Here, height (z) refers to, for example, the error in the height direction when moving up and down with an overhead crane. Error consideration (perpendicular to the direction of travel), error consideration (direction of travel), and error consideration (z) are flag information that determines whether the error estimated by the error estimation unit 111 is considered in the direction perpendicular to the direction of travel of the 2D specific path (a specific path in a 2D plane), the direction of travel of the 2D specific path, and the 3D direction, respectively, in the 3D map processing unit 112 and the extended path creation unit 113.

[0036] In specific example 402, if error consideration (perpendicular to the direction of travel) is set to "True", the extended route creation unit 113 takes into account the error estimated by the error estimation unit 111 (described later). If error consideration (direction of travel) and error consideration (z) are both set to "True", the 3D map processing unit 112 adds the error estimated by the error estimation unit 111 (described later) to the width and height of obstacles on the 3D map.

[0037] Furthermore, if error consideration (perpendicular to the direction of travel) is set to "False", the extended route creation unit 113 will not consider errors in the direction perpendicular to the direction of travel when creating the extended transport route information 207, even if the error information is finite. If at least one of error consideration (direction of travel) and error consideration (z) is set to "False", the 3D map processing unit 112 will not consider errors in the direction of "False" (direction of travel, z-direction) when creating the processed 3D map 205.

[0038] (Extended transport route information 207) The extended transport route information 207 is an extended two-dimensional route created by the extended route creation unit 113. The extended two-dimensional route is obtained by adding the width (width in a two-dimensional plane) created by the extended route creation unit 113 based on the error information 201 and the route width 206 to the transport route information 202. In the specific example 403 shown in Figure 8, width (left) and width (right) are added to the specific example 303 of the transport route information 202. Width (left) and width (right) are the widths in the left-right direction of the central axis when the route in the two-dimensional plane is considered as the central axis. For example, consider the case where width (left) and width (right) are determined based on the two-dimensional route numerical error (fixed) and two-dimensional route numerical error (time) described in the error information 201 shown in Figure 5. In the example in Figure 5, the x direction and y direction are assumed to be the same, so width (left) and width (right) are calculated by converting these numerical errors in the x and y directions into numerical errors in the width direction with the route as the central axis.

[0039] (3D transport information 208) The 3D transport information 208 is information such as the transport position of the moving object and the transported object at each time, as searched by the transport path search unit 114. Specific example 404 shown in Figure 9 is a specific example of the 3D transport information 208. The 3D transport information 208 includes the ID of the moving object, time, the position (x), position (y), position (z), position (x), position (y), and position (z) of the transported object. In addition to the information shown in specific example 404, it may also include the velocity, swing angle, acceleration, and deceleration of the moving object and the transported object at each time, as well as information regarding the acceleration and deceleration control of the moving object and the transported object. Note that the unit of position is not limited to [mm], but may also be [m], etc.

[0040] (3D map 204) The 3D map 204 is a two-dimensional planar map that includes height information of obstacles at each point on a field where moving objects can move. For example, the 3D map 204 may have multiple grids with fixed lengths in the x and y directions within a two-dimensional plane, and each grid may contain information such as the highest or average height of obstacles within that grid. By combining this information, the 3D map 204 represents the size of obstacles in three-dimensional space.

[0041] The grid width can be chosen arbitrarily. Coordinate systems like those in Euclidean space can be considered as grids with infinitesimal widths in numerical calculations, and therefore can be viewed as grids like 3D map 204. In other words, by setting the grid width to a very small value, it can be viewed as a map with continuously changing heights, similar to Euclidean space.

[0042] (Specific example of 3D map 204) Figure 10 shows a specific example 501 of the 3D map 204. In example 501, grid 502 is set as the starting position of the grid, and the coordinates (x, y) of grid 502 are set to (x, y) = (0 mm, 0 mm), but other settings are also acceptable. The numerical values ​​written within the grid represent the highest height of the obstacles, but they can also be average values ​​or any other value that represents the height information of the obstacles. In example 501, the numerical values ​​are expressed in [mm] units, but they can also be expressed in [m] units, and the units are not limited. In example 501, obstacles exist in the areas indicated by the dashed lines L1 and L2. These obstacle areas L1 and L2 may each represent a single obstacle, or they may consist of multiple obstacles.

[0043] (Processed 3D map 205) The processed 3D map 205 is a 3D map processed by the 3D map processing unit 112. The processed 3D map 205 is obtained by modifying the height information of obstacles and the width in the x and y directions compared to the 3D map 204, taking into account the 3D map processing width 203, the path width 206, and the error estimated by the error estimation unit 111.

[0044] (Specific example of processed 3D map 205) Figure 11 shows a specific example 503 of the processed 3D map 205. Specific example 503 is based on specific example 401 of the 3D map 204 shown in Figure 10, but it is an example where no width is added in the x-direction for the 3D map processing width 203 shown in Figure 7. In Figure 11, the obstacle range L1 is expanded to the positive and negative sides of the y-direction. The obstacle range L2 is also expanded to the positive and negative sides of the y-direction. Furthermore, the height of the obstacles in each grid included in ranges L1 and L2 in Figure 11 is increased by 500 mm. In grid 504, the expanded ranges L1 and L2 overlap, but in this case, the height value is the height of the taller obstacle. Note that the obstacle height in grid 504 may be the average value or the lower height.

[0045] (3D transport distance information 209) The 3D transport distance information 209 includes the x- and y-direction travel distances of the moving object when the 3D transport information 208 is obtained, as well as the z-direction upward and downward distances of the moving object and the transported object. The specific example 601 shown in Figure 12 is a specific example of the 3D transport distance information 209. The 3D transport distance information 209 includes the ID of the moving object, the time, the travel distance of the moving object (x-direction), the travel distance of the moving object (y-direction), the travel distance of the moving object (z-direction upward), the travel distance of the moving object (z-direction downward), the travel distance of the transported object (z-direction upward), the travel distance of the transported object (z-direction downward), the control pattern number for the upward and downward movement of the transported object, and detailed control.

[0046] The control pattern number for raising and lowering the transported object is a number that represents the control pattern used by the moving object to raise and lower the transported object. The control patterns for raising and lowering the transported object are pre-configured, and when the transport path search unit 114 (described later) determines the appropriate control pattern number, that control pattern number is stored. If there is no control pattern number for raising and lowering the transported object, the system will indicate that there is no information by displaying "-" or "None".

[0047] In the detailed control section, if the control pattern number for the upward and downward movement of the conveyed object is listed, it should be marked with "-" or left blank. If the control pattern number for the upward and downward movement of the conveyed object is not listed, then, as shown in Specific Example 601, time:acceleration or time:deceleration should be listed corresponding to each time point. In Specific Example 601, the unit of distance is [mm], but it is not limited to this and may also be [m].

[0048] (3D transport path 210) The 3D transport path 210 stores information about the 3D transport path, including the path considered by the transport path search unit 114 (described later), the path in a 2D plane that takes errors into account (extended 2D path), and the transport in the 3D direction.

[0049] (Route creation method 211) Route creation method 211 describes a transport method in which the route creation unit 110 creates a route (a route in a two-dimensional plane in the x and y directions) when the transport route information 202 does not contain a transport start position and a transport end position. Specific example 602 in Figure 13 shows a specific example of route creation method 211. Route creation method 211 may describe transport patterns such as performing theoretically shortest transport in the x and y directions, or it may describe other transport methods.

[0050] (Ascent / Descent Control Pattern 212) The ascent / descent control pattern 212 describes the control pattern for the ascent and descent of the transported object when the transport path search unit 114 performs a search. Specific example 603 in Figure 13 shows a specific example of the ascent / descent control pattern when applied to an overhead crane, and includes the ID of the moving object and the winding pattern, which is the method of ascent and descent of the transported object. In the case of a drone, the ascent / descent pattern of the drone holding the transported object is described.

[0051] The upward / downward control pattern may be expressed in natural language as in Specific Example 603, simply as a number, or as a specific control method as in the detailed control in Specific Example 602. If expressed in natural language, the transport path search unit 114 performs the control corresponding to this natural language. If the winding pattern is not expressed, a display indicating that there is no information, such as "-" or "None," will be shown.

[0052] <Detailed explanation of processing in the route search unit 103> Next, we will explain the processing performed in each part of the route search unit 103. Figure 14 is a flowchart showing the processing procedure in the route search unit 103.

[0053] In step S801, the route search unit 103 obtains information on a specific route (the transport route information listed in the row corresponding to the ID of the object being moved) from among the routes in the two-dimensional plane described in the transport route information 202 of the database 102.

[0054] In step S802, the route search unit 103 determines whether the information included in the acquired specific route consists only of the transport start position and the transport end position. If the result in step S802 is YES (affirmative), the process proceeds to step S803; if the result is NO (negative), the process proceeds to step S804.

[0055] For example, in the case of the transport route information 202 described in specific example 303, the example with ID=0 includes time information other than the transport start position (time=0s) and transport end position (time=25.2s), so it is determined to be NO in step S802. On the other hand, in the example with ID=1, only the transport start position and transport end position information is described, so it is determined to be YES in step S802.

[0056] In step S803, the path search unit 103 causes the path creation unit 110 to create a specific path in a two-dimensional plane.

[0057] (Detailed processing of step S803) Figure 15 is a flowchart showing the detailed processing of step S803 (processing by the route creation unit 110). In step S8031, the route creation unit 110 obtains the moving object information 200, the route creation method 211, and the transport route information 202 from the database 102 of the data storage device 702. In this case, the transport route information 202 contains only the information of the transport start position and the transport end position.

[0058] In step S8032, the route creation unit 110 creates a specific route connecting the transport start position and the transport end position in a two-dimensional plane, based on the transport method described in the route creation method 211 and the information such as the maximum speed and acceleration described in the moving object information 200.

[0059] As an example of the route creation method 211, specific example 602 describes a transport method that "transports using the theoretically shortest transport method for both the x and y directions." For example, when this transport method is applied, step S8032 creates a route connecting the transport start position and the transport end position so as to transport using this theoretically shortest transport method. Other examples of route creation methods include, for example, a transport method that transports at a constant speed in both the x and y directions, or a transport method in which acceleration and deceleration information at each time point, learned by machine learning, is described as time-series information, and the transport method is not limited.

[0060] In step S8033, the route creation unit 110 stores the route created in step S8032 as a specific route in the transport route information 202.

[0061] Returning to the flowchart in Figure 14, in step S804, the path search unit 103 is instructed to perform error estimation by the error estimation unit 111.

[0062] (Detailed processing of step S804) Figure 16 is a flowchart showing the detailed processing of step S804 (processing by the error estimation unit 111). In step S8041, the error estimation unit 111 acquires the moving object information 200, the error information 201, and information about a specific route within the transport route information 202. The information about a specific route acquired here is information about the specific route acquired in step S801, or the specific route created in step S803.

[0063] In step S8042, the error estimation unit 111 estimates the errors at each time point for the 2D path numerical error and the 3D direction numerical error, respectively, based on the 2D path numerical error (fixed), 2D path numerical error (time), 3D direction numerical error (fixed), and 3D direction numerical error (time) of the acquired error information 201.

[0064] One example of an error estimation method in step S8042 is to add a time-varying error to a fixed error at each time step. However, addition is not required; arithmetic operations are also acceptable, and the error can be estimated by inputting it as an argument to a function such as an exponential function. Furthermore, the error can be estimated using machine learning or other methods, and the error estimation method is not limited.

[0065] In step S8043, the error estimation unit 111 uses the errors calculated in step S8042 (the numerical errors of the 2D path and the numerical errors of the 3D direction at each time point) to estimate the errors necessary for creating the processed 3D map from the error in the direction of travel of the 2D path, the numerical error in the 3D direction, and the errors related to the 3D map (3D map sensor errors and 3D map information errors in the x, y, and height directions). The errors necessary for creating the processed 3D map also include the error that the height of obstacles placed within the map on which moving objects can move may have changed since the creation of the 3D map 204.

[0066] Furthermore, the error in the direction of travel when a moving object travels along a specific path in a two-dimensional plane is estimated based on the width and size of the transported object, the maximum speed and acceleration information of the moving object included in the moving object information 200, and the error information 201.

[0067] In step S8044, the error estimation unit 111 uses the errors calculated in step S8042 (the errors at each time for the 2D path numerical error and the 3D direction numerical error) to estimate the error required when creating the extended 2D path from the error in the direction perpendicular to the direction of travel of the 2D path. The error corresponding to the width perpendicular to a specific path is estimated based on the width and size of the transported object included in the moving object information 200, the maximum speed and acceleration information of the moving object, and the error information 201.

[0068] In step S8043 and S8044, the error estimation can be performed by adding each error equally, adding them with weights based on their magnitude, using arithmetic operations instead of addition, or estimating the error by inputting it as an argument to a function such as an exponential function. Furthermore, the error can be estimated using machine learning or other methods, and the method of error estimation is not limited.

[0069] Returning to the flowchart in Figure 14, in step S805, the route search unit 103 causes the 3D map processing unit 112 to create a processed 3D map.

[0070] (Detailed processing of step S805) Figure 17 is a flowchart showing the detailed processing of step S805 (processing by the 3D map processing unit 112). In step S8051, the 3D map processing unit 112 obtains the "error required when creating the processed 3D map" estimated by the error estimation unit 111 from the error estimation unit 111, and also obtains the 3D map processing width 203 and the path width 206 from the database 102.

[0071] The information obtained from the path width 206 includes the width in the direction of travel and the z-direction, as well as error considerations. The width in the direction of travel and the z-direction are taken into account as the height and width of obstacles present on the map and reflected in the 3D map 204.

[0072] In step S8052, the 3D map processing unit 112 processes the 3D map 204 based on the error consideration information described in the 3D map processing width 203 and path width 206 to create a processed 3D map 205. That is, for directions where it is stated that no error should be considered in the 3D map processing width 203 and path width 206, the 3D map 204 is processed based on the user-specified value of the 3D processing width. On the other hand, if it is stated that an error should be considered, the 3D map 204 is processed based on the error based on the error information 201 and the width specified by the user. That is, a processed 3D map 205 is created by adding the acquired error, width, and height to the width in the 2D plane and height in the 3D direction of the obstacles in the 3D map 204.

[0073] When considering errors, the error and the user-specified width may be added equally, weighted based on their magnitude, or calculated using arithmetic operations instead of simple addition. The width may also be calculated by inputting the error as an argument into a function such as an exponential function. Furthermore, the width may be estimated using machine learning or other methods, and the method of estimating the width is not limited.

[0074] Furthermore, regarding the width added by the 3D map processing unit 112, the amplitude of the conveyed object's movement may be estimated and added to the width added to the 3D map 204. The amplitude of the conveyed object's movement is estimated, for example, based on the width and height of the conveyed object, the velocity and acceleration information of the moving object included in the moving object information 200, and the conveying path information 202 in the 2D plane. Alternatively, the conveying path information 202 shown in Figure 6 may include information on the amplitude of movement.

[0075] In step S8053, the 3D map processing unit 112 stores the processed 3D map in the database of processed 3D maps 205.

[0076] Returning to the flowchart in Figure 14, in step S806, the route search unit 103 causes the extended route creation unit 113 to create extended transport route information (extended 2D route).

[0077] (Detailed processing of step S806) Figure 18 is a flowchart showing the detailed processing of step S806 (processing by the extended route creation unit 113). In step S8061, the extended route creation unit 113 obtains the error for the estimated specific route from the error estimation unit 111 and also obtains the route width 206 from the database 102.

[0078] In step S8062, the extended route creation unit 113 creates an extended two-dimensional route based on the error consideration information described in the route width 206. If the error consideration information states "do not consider error," the extended two-dimensional route is created based on the user-specified route width value. On the other hand, if the error consideration information states "consider error," the extended two-dimensional route is created based on the error of the error estimation unit 111 and the width specified by the user.

[0079] When considering errors, the error and the user-specified width can be added equally, weighted based on their magnitude, or even performed using arithmetic operations instead of addition. The width can also be estimated by inputting the error as an argument to a function such as an exponential function. Furthermore, the width can be estimated using machine learning or other methods, and the method of estimating the width is not limited.

[0080] In step S8063, the extended route creation unit 113 stores the extended two-dimensional route created in step S8062 in the extended transport route information 207.

[0081] Returning to the flowchart in Figure 14, in step S807, the path search unit 103 causes the transport path search unit 114 to create 3D transport information.

[0082] (Detailed processing of step S807) Figure 19 is a flowchart showing the detailed processing of step S807 (processing by the transport path search unit 114). In step S8071, the transport path search unit 114 acquires the moving object information 200, the processing 3D map 205, the extended transport path information 207, and the up / down control pattern 212.

[0083] In step S8072, the transport path search unit 114 determines whether or not a winding pattern is described in the upward / downward control pattern 212. If a winding pattern is described in the upward / downward control pattern 212, the unit proceeds to step S8703; otherwise, the unit proceeds to step S8704.

[0084] In step S8073, the transport path search unit 114 determines a winding pattern that, when the upward and downward control using the winding pattern described in the upward and downward control pattern 212 is performed on the extended two-dimensional path created in step S806, will not collide with any obstacles that overlap and interfere with the extended two-dimensional path on the processing three-dimensional map 205, and will also minimize the overall transport time. Once the processing in step S8073 is complete, the process proceeds to step S8075.

[0085] In step S8074, the transport path search unit 114 searches for a winding pattern that does not collide with obstacles overlapping and interfering with the extended 2D path on the processing 3D map 205, and that minimizes the overall transport time, based on the moving object information 200. Once the processing in step S8074 is complete, the process proceeds to step S8075.

[0086] The process proceeds from step S8072 to step S8074 only if no winding pattern is specified in the upward / downward control pattern 212 in step S8072. If no winding pattern is specified, the system may explore whether a predetermined pattern is feasible, such as "winding the transported object up to the highest point of the obstacle before transport begins, and winding the transported object down to a specified height after transport is complete." Alternatively, the system may explore transport in three dimensions by performing multiple searches, such as "raising the transported object as the moving object moves, and if a collision occurs with an obstacle, restarting the search from the beginning, and evaluating whether the transported object will collide with an obstacle after raising it a certain number of times before the moving object moves."

[0087] Similarly, for the descent of the transported object, one method is to "evaluate whether the moving object will collide with the transported object while it is moving, try multiple methods such as changing the timing of the transported object's unwinding, or lowering the transported object at the end of transport, and select the transport method that results in the shortest transport time," or any other method is also acceptable.

[0088] Furthermore, in the process of step S8073 described above, if all of the winding patterns described in the upward / downward control pattern 212 collide with an obstacle, a three-dimensional transport search (search for a path and upward / downward patterns) as described in the process of step S8074 may be performed.

[0089] In step S8075, the transport path search unit 114 creates 3D transport information based on the transport information in the searched 3D direction, using the position, acceleration, deceleration, and velocity information of the moving object in each direction in the 2D plane at each time point, and the position, acceleration, deceleration, and velocity information in the 3D direction. It then outputs the created 3D transport information and the winding pattern number or the method for controlling the raising and lowering of the transport object. The created 3D transport information is also stored in the 3D transport information 208 of the database 102.

[0090] Returning to the flowchart in Figure 14, in step S808, the path search unit 103 causes the output result creation unit 115 to output 3D transport path information.

[0091] (Detailed processing of step S808) Figure 20 is a flowchart showing the detailed processing of step S808 (processing by the output result creation unit 115). In step S8081, the output result creation unit 115 acquires the 3D transport information obtained by the transport path search unit 114 and the up / down control pattern (winding pattern number or method for controlling the up / down movement of the transported object).

[0092] In step S8082, the output result creation unit 115 calculates the distance traveled by the moving object and the transported object in each direction, as well as the upward and downward control of the transported object (the number if controlled by a winding pattern number), based on the acquired 3D transport information and upward and downward control pattern, and stores them in the 3D transport distance information 209.

[0093] In step S8083, the output result creation unit 115 creates path information of the transported object based on the acquired 3D transport information and the up / down control pattern. For example, it creates a movement diagram in a 2D plane and a movement diagram in a 3D direction. The created path information is then stored in the 3D transport path 210 and output. For example, it is output to the output device 705.

[0094] Furthermore, the path information stored in the 3D transport path 210 does not have to be a movement diagram in a 2D plane and in a 3D direction as described above. For example, it may be a diagram of movement in 3D space, or it may be a diagram illustrating the paths in the x, y, and z directions. Also, the path information does not have to be movement diagram information, but may be numerical information (3D transport information and 3D transport distance information) that formed the basis of the movement diagram. In any case, this information is path information consisting of transport position information and transport distance information.

[0095] The flowchart above illustrates the case where only one ID is set for the moving object, i.e., when there is only one moving object. However, if two or more IDs are set and transportation actually takes place simultaneously, each step in the flowchart above should be performed for each ID. The 3D transport movement diagram in step S8083 may be created separately for each ID, or the path information for each ID may be illustrated in a single diagram.

[0096] <Example 1: Conveyance using an overhead crane> Figures 21-25 show an example in which the pathfinding device 100 of this embodiment is used for transport by an overhead crane. Figure 21 is a schematic diagram showing the general outline of overhead crane transport. 1501 is the trolley of the overhead crane, and the trolley 1501 is movable in the x and y directions in the horizontal plane. The trolley 1501 transports the transported object 1502 suspended by a wire 1503. In Embodiment 1, the trolley 1501 corresponds to the moving object, and the positions of the moving object (trolley 1501) and the transported object 1502 are different.

[0097] The rectangles labeled 1505a, 1505b, and 1505c represent obstacles on the field 1504, and the overhead crane's trolley 1501 can move across the field 1504. The label S represents the transport start position, and the label G represents the transport end position. The trolley 1501 transports the suspended object 1502 from the transport start position S to the transport end position G, taking care not to collide with the obstacles 1505a to 1505c.

[0098] In Example 1, the transport route information 202 is a route in which only the transport start position S and the transport end position G are determined, as shown in Example 303 with ID=1. Therefore, the route creation unit 110 creates a specific route in a two-dimensional plane. The route creation method 211 is to perform theoretically shortest transport in the x and y directions, as specified with ID=0 in Example 602.

[0099] The error estimation unit 111 performs error estimation assuming that the error information 201 is an error that includes time-varying numerical errors, as in the case of ID=0 in specific example 302, and that the error consideration for the 3D map processing width 203 and path width 206 is "True" in all directions.

[0100] The 3D map processing unit 112 creates a processed 3D map 205 by adding width and height in a 2D plane to the 3D map 204, as in the example 503, based on the error estimated by the error estimation unit 111. However, unlike the example 503, in Example 1, width is also given in the y direction.

[0101] Figure 22 is a schematic diagram showing the processed 3D map 205 as a 2D map viewed from the positive z direction, with a 2D path drawn on that 2D map. Field 1510 is a 2D representation of field 1504. The dashed rectangles indicated by symbols 1509a to 1509c are obstacles on the processed 3D map 205, and are processed by the 3D map processing unit 112 from obstacles 1505a to 1505c on the 3D map 204 shown in Figure 21. As a result of the processing of obstacles 1505a to 1505c by the 3D map processing unit 112, obstacles 1509a to 1509c have increased in width.

[0102] Note that the actual 3D processing map 205 has a grid structure as shown in specific example 503, and height information is recorded for each grid. However, in Figure 22, this numerical information is omitted, and the grid is omitted for simplicity. Of course, a diagram showing both the grid and height is also acceptable.

[0103] The line with an arrow labeled 1511 represents the path before expansion (a specific path in a two-dimensional plane), and the direction of the arrow indicates the direction of travel along the path. Of course, such arrows are not required for an actual path. The moving object (trolley 1501) moves along this path 1511. The expanded path creation unit 113 adds width perpendicular to the direction of travel of the path 1511 in the two-dimensional plane of Figure 22, based on the error estimated by the error estimation unit 111. The pair of lines labeled 1512 represent the width added by the expanded path creation unit 113. As the transport progresses from the transport start position S to the transport end position G, the width between the lines 1512 increases depending on the time change of the numerical error.

[0104] The transport path search unit 114 uses the information of the processed 3D map 205, schematically shown in Figure 22, and the extended path (extended 2D path) to perform a 3D path search. The extended 2D path is the region enclosed by a pair of lines 1512 indicating the width. Here, it is assumed that the same winding pattern as in the case of ID=0 described in specific example 603 is described as the upward / downward control pattern 212. Using this upward / downward control pattern 212, a 3D path search is performed to ensure that the transported object 1502 (Figure 21) does not collide with obstacles 1509b and 1509c that are described overlapping the extended path.

[0105] The upward and downward movement distance of the conveyed object 1502 is estimated by adding the information on the height of the conveyed object at the conveying start position S and conveying end position G, as described in Specific Example 403, the difference in the maximum heights of obstacles 1509b and 1509c located within the area between the lines 1512, which are extended two-dimensional paths, and a distance that takes into account the size of the conveyed object 1502. In Example 1, the height of the conveyed object at the conveying start position S and conveying end position G is assumed to be the same as in the case of ID=1 in Specific Example 403. Therefore, if the maximum height of the obstacle is 8000 mm, the upward movement is 2000 mm and the downward movement is also 2000 mm. In Example 1, the same winding pattern as winding pattern 1 in ID=0 of Specific Example 603 is selected as the upward / downward control pattern 212.

[0106] In Example 1, based on the above-described ascension / descension control pattern 212 and the height information of the transported object and obstacles at the transport start position S and transport end position G, the transport path search unit 114 outputs 3D transport information similar to that in the case of ID=0 in Specific Example 404, and the ascension / descension control pattern number determined by the search.

[0107] The output result creation unit 115 outputs three-dimensional transport distance information 1521, as shown in Figure 23, and three-dimensional transport paths, as shown in Figures 24 and 25, from the three-dimensional transport information output from the transport path search unit 114 and the ascent / descent control pattern number determined by the search.

[0108] Figure 24 shows an example of output example 1601 of a path in a two-dimensional plane. It shows a symbol S indicating the start position of transport and a symbol G indicating the end position of transport. Obstacles 1609b and 1609c are shown overlapping the extended two-dimensional path 1604, which includes errors. Obstacles 1609a to 1609c correspond to obstacles 1509a to 1509c on the processed three-dimensional map 205, but obstacles 1505a to 1505c on the three-dimensional map 204 may be displayed instead. Also, the display for the extended two-dimensional path 1604 can be anything that shows the path of the moving object, for example, the path before extension 1511 (Figure 22) may be used.

[0109] Output example 1601 in Figure 24 is just one example; numerical values ​​may be included on the axes of the figure, height information for each obstacle may be included on the map, or the height difference between the transported object and the obstacles may be included. Furthermore, the transport start position S and transport end position G do not need to be included. In addition, output example 1601 may be represented as a color plot based on the height difference between the transported object and the obstacles.

[0110] The output example 1611 shown in Figure 25 represents the direction of travel of the moving object (trolley 1501) on the extended two-dimensional path 1604, with the horizontal axis of the figure and the height in the three-dimensional direction on the vertical axis. The line labeled 1612 is the path of the transported object 1502, and the rectangles labeled 1613b and 1613c schematically represent the heights of the obstacles 1509b and 1509c. The path 1612 of the transported object 1502 may be a line without width, as shown in Figure 25, or it may be displayed with a width.

[0111] Output example 1611 in Figure 25 is just one example; numerical values ​​may be indicated on the vertical and horizontal axes of the figure, arrows indicating the direction of travel may be indicated on the route 1612, and the transport start position S and transport end position G may be indicated. Furthermore, the moving object (trolley 1501) may be shown in the figure, and a specific drawing of the transported object 1502 may be shown. Moreover, the display format does not have to be as shown in output example 1611, as long as the path in which the transported object overcomes obstacles is indicated.

[0112] After confirming the 3D transport distance information 1521 and the 3D transport path in output example 1601 and output example 1611, the user determines the control method for the x and y directions of the overhead crane and controls the overhead crane while hoisting up the transported object 1502 based on the up and down control pattern described in the 3D transport distance information 1521.

[0113] Furthermore, the user may omit the verification of the 3D transport distance information 1521 and the 3D transport path in output examples 1601 and 1611, and if a path has been found, the overhead crane may be automatically operated based on the transport distance in the 3D transport distance information 1521 and the information on the upward and downward control pattern.

[0114] <Example 2: Transport by drone> Figures 26-29 show an example in which the pathfinding device 100 of this embodiment is used for transport by drone. Figure 26 is a schematic diagram showing the outline of transport by drone. In Example 2, the 3D map is not a grid map as described in Specific Example 501, but a 3D map 1710 represented in continuous space. As shown in Figure 26, each axis of the 3D spatial coordinates (x, y, z) is set, and each coordinate is a continuous value.

[0115] The curved surface indicated by reference numeral 1712 represents an obstacle present on the field and represents the height of the obstacle 1712. Reference numeral 1700a represents a moving object consisting of a drone 1701 and a transported object 1702. Since the transported object 1702 is in contact with the drone 1701, which is the original moving object, in Example 2, the drone 1701 and the transported object 1702 are treated together as moving object 1700a. In other words, Example 2 is an example where the item "Are the transported object and the moving object the same?" is "True", as in the specific example 301 of the moving object information 200 with ID=1.

[0116] In Example 2, as shown in Figure 26, there are two moving objects 1700a and 1700b (moving object IDs = 0 and 1), representing an example where there are multiple moving objects. The positions indicated by symbols S0 and G0 are the start and end positions of transport for moving object ID = 0, and the positions indicated by symbols S1 and G1 are the start and end positions of transport for moving object ID = 1.

[0117] The two-dimensional map 1721 shown in Figure 27 is a projection of the three-dimensional map 1710 shown in Figure 26 onto the xy plane. Height information is omitted. Example 2 describes the case where a route other than the transport start position and transport end position is specified as the transport route information 202. Obstacle 1722 is a projection of obstacle 1712 shown in Figure 26 onto the xy plane. Furthermore, the same reference numerals are used for the projections of the transport start positions S0, S1 and the transport end positions G0, G1 onto the xy plane.

[0118] Route 1727 is the route connecting the transport start position S0 and the transport end position G0, and route 1728 is the route connecting the transport start position S1 and the transport end position G1. Route 1728 is a route obtained by machine learning or other methods using a separate device, and may be a route determined by acceleration and deceleration control at each time in the x and y directions, as shown in the detailed control of specific example 602 (route creation method 211).

[0119] As described above, since the paths for both moving object ID=0 and 1 have already been determined, step S802 in the flowchart of Figure 14 is determined to be NO, and the error estimation process by the error estimation unit 111 in step S804 is executed. Here, in the error estimation unit 111 of Example 2, each piece of error information 201 is unclear, and it is assumed that the error cannot be estimated accurately. However, it is assumed that the user has performed drone transport many times in the past and has an understanding from experience of the errors between the transport position when transporting along a specific path and the errors when creating the 3D map.

[0120] The user inputs the error between the route known from experience and the 3D map into the widths of the 3D map processing width 203 and the route width 206. Furthermore, the flag information (error consideration (x), error consideration (y), and error consideration (z)) listed in the 3D map processing width 203 and the route width 206 is set to "False" for all directions. In this state of the 3D map processing width 203 and route width 206, it is difficult to estimate numerical errors, but by applying the width and height that take into account the errors obtained from the user's knowledge to the 3D map 204, it is possible to create the processed 3D map 205 by the 3D map processing unit 112, and to create the extended route (corresponding to an extended 2D route) described later by the extended route creation unit 113.

[0121] In Example 2, we assumed that the user had empirical knowledge about errors. However, even if the user does not have such knowledge, it is possible to create a processed 3D map 205 and an extended 2D path that take into account possible errors by setting a width that the user deems safe.

[0122] The transport path search unit 114 uses the processed 3D map 205 and the extended 2D path described above to search for an ascent / descent path for each moving object so that the moving object can overcome the height of obstacles on the extended 2D path. In Example 2, it is assumed that no winding pattern (ascent / descent pattern) is specified for the ascent / descent control pattern, as in the case of ID=1 in Specific Example 603. In this case, step S8072 of the flowchart in Figure 19 is determined to be NO, the search is performed comprehensively in step S8074, and the 3D transport information after the search and the control method for the ascent / descent of the moving object in the 3D direction obtained in the search are output in step S8075.

[0123] The output result creation unit 115 outputs 3D transport distance information 1730, as shown in Figure 28, and information regarding the 3D transport path, as shown in Figure 29, based on the outputted 3D transport information and the control method for raising and lowering the moving object in the 3D direction. The detailed control of the 3D transport distance information 1730 shown in Figure 28 describes whether acceleration (accel) and deceleration (brake) were selected at each time point.

[0124] Figure 29 shows a processed 3D map obtained by processing the 3D map shown in Figure 26 with the 3D map processing unit 112. In Figure 29, the 3D paths 1727 and 1728 of moving objects 1700a and 1700b are also superimposed on the processed 3D map. The curved surface labeled 1802 represents an obstacle in the processed 3D map and is a curved surface that represents the height of the obstacle. The obstacle 1802 in the processed 3D map is created by processing the obstacle 1712 on the 3D map with the 3D map processing unit 112.

[0125] A planar extension path 1808 having a horizontal width is set for the path 1727 connecting the transport start position S0 and the transport end position G0. A planar extension path 1810 having a horizontal width is set for the path 1728 connecting the transport start position S1 and the transport end position G1.

[0126] In the case of an overhead crane, the moving object (trolley 1501) can only move in the x and y directions, so we considered an extended path in a two-dimensional plane (extended two-dimensional path) relative to the moving object. However, in the case of a drone, the moving object can move in three dimensions, so we consider a three-dimensional extended path, as shown in extended paths 1808 and 1810. The extended two-dimensional path is obtained by projecting extended paths 1808 and 1810 onto the xy plane.

[0127] Note that the extension paths 1808 and 1810 shown in Figure 29 are just examples. The extension paths 1808 and 1810 may also be represented as tubes, or, as in Example 1, the extension paths with a two-dimensional plane and height as axes may be described for each moving object ID, or the two IDs may be described together.

[0128] After the user confirms the 3D transport distance information 1730 (Figure 28) and the 3D transport path in Figure 29, they determine the control method for the x and y directions for object ID=0. For object ID=1, they use the control method determined by machine learning. The user then controls the drone based on the ascent and descent control of the object described in the detailed control of the 3D transport distance information 1730. Alternatively, the user may omit confirming the 3D transport distance information 1730 and the 3D transport path in Figure 29, and after the path has been found, automatically operate the drone based on the transport distance and ascent / descent control pattern information from the 3D transport distance information 1730.

[0129] The embodiments and modifications of the present invention described above provide the following effects.

[0130] (C1) As shown in Figures 1, 8, 21, 22, 24, etc., the 3D transport path search device 100 searches for a 3D transport path for the trolley (moving object) 1501 that transports the transported object 1502. The system includes a database (storage unit) 102 that stores extended transport path information (extended travel path) 207 obtained by adding a first width based on the position error of the trolley 1501 (for example, included in the width (left) and width (right) described in the extended transport path information 207 in Figure 8, and the width between the pair of lines 1512 in Figure 22) to the path (movement path) 1511 of the trolley 1501, and a 3D map (3D map information) 204 that includes information on the size of obstacles 1505a to 1505c, and a transport path search unit 114 that generates a 3D transport path in which the trolley 1501 and the transported object 1502 do not collide with obstacles 1505a to 1505c based on the extended transport path information 207 and the 3D map 204 stored in the database 102.

[0131] As described above, the transport path search unit 114 sets an extended two-dimensional path 1604 obtained by assigning a first width based on the position error of the trolley (moving object) 1501, and generates a three-dimensional transport path based on the extended two-dimensional path 1604 in which the trolley 1501 and the transported object 1502 do not collide with obstacles 1609b and 1609c. In this way, by considering errors related to the moving object (position error of the moving object) when generating a three-dimensional transport path that can avoid collisions, it is possible to search for a highly accurate three-dimensional transport path. In particular, time-dependent errors (for example, two-dimensional path numerical error (time)) described in the error information 201 of Figure 5 can be easily applied by using the extended two-dimensional path 1604.

[0132] In the above-described embodiment, an example is explained in which a search is performed using a processed 3D map 205 that takes into account 3D map sensor errors, etc. However, even when searching for a 3D transport path using the 3D map 204 before processing, by taking into account errors related to the moving object, it is possible to search for a 3D transport path with higher accuracy than conventional methods.

[0133] Furthermore, as in Example 2, when transport is performed by the drone 1701, the extended path 1808 shown in Figure 29 corresponds to the extended movement path, and a highly accurate 3D transport path can be searched, similar to Example 1. In Figure 29, even when the 3D transport path is searched using obstacles 1712 on the 3D map instead of obstacles 1802 on the processing 3D map, a more accurate 3D transport path can be searched compared to conventional methods by taking into account errors related to the moving object.

[0134] The 3D map (3D map information) 204 is, for example, a 2D plane gridded with a finite grid, as shown in Figure 10, and includes height information of obstacles present within the grid. However, the width of the grid can be arbitrarily chosen, and by setting the width to a very small value, it can be considered a map with continuously changing heights, similar to Euclidean space. In other words, the 3D map 204 is not limited to a gridded map, but can also represent a continuous space.

[0135] (C2) In (C1) above, as shown in Figures 5, 22, etc., the first width described above is calculated by converting the horizontal position error of the movement path (1511) of the moving object (trolley 1501) (the two-dimensional path numerical error (fixed) and the two-dimensional path numerical error (time) shown in error information 201 in Figure 5) into a horizontal width with the movement path (1511) as the central axis.

[0136] (C3) In (C1) above, as shown in Figures 1, 5, 10, 11, etc., the system further includes a 3D map processing unit 112 that adds a second width to the information on the size of obstacles included in the 3D map (3D map information) 204, based on at least one of the errors that occur when creating the 3D map 204 (for example, the 3D map sensor error shown in Figure 5), the amplitude of the transported object, and the position error in the direction of travel when the moving object moves along the extended travel path, and generates a processed 3D map (processed 3D map information) 205. The transport path search unit 114 generates a 3D transport path based on the processed 3D map 205 instead of the 3D map 204.

[0137] As described above, by reflecting the errors that occur when creating the 3D map 204, the amplitude of the transported object's movement, and the positional error in the direction of travel when the moving object proceeds along the extended movement path in the form of a second width in the size information of the obstacle, it becomes possible to search for a transport path that can avoid collisions between the transported object and obstacles due to the effects of the above errors and the amplitude of the transported object's movement.

[0138] (C4) In (C1) above, as shown in Figures 19, 22, 24, etc., the extended movement path is an extended two-dimensional path (extended two-dimensional movement path) 1604 on a two-dimensional plane, and the transport path search unit 114 searches for an ascent / descent control pattern that does not collide with obstacles 1609b, 1609c interfering with the extended two-dimensional path 1604 when the transport of the moving object is carried out on the extended two-dimensional path 1604 from the transport start position S to the transport end position G, and that minimizes the overall transport time, thereby generating a three-dimensional transport path.

[0139] By setting an extended two-dimensional path 1604 for obstacles 1609b and 1609c and searching for an ascent / descent control pattern, it is possible to evaluate collisions with obstacles while considering errors, and to generate a three-dimensional transport path that can reliably avoid collisions with obstacles.

[0140] (C5) In (C4) above, as shown in Figures 13, 19, 24, etc., the up / down control pattern 212 of the database 102 further stores multiple winding patterns (up / down control patterns). The transport path search unit 114 selects from the multiple up / down control patterns stored in the up / down control pattern 212 that, when transporting the moving object along the extended two-dimensional path (extended two-dimensional movement path) 1604 from the transport start position S to the transport end position G, will not collide with obstacles 1609b, 1609c interfering with the extended two-dimensional path 1604 and will result in the shortest overall transport time, and generates a three-dimensional transport path.

[0141] By exploring multiple ascent and descent control patterns, a more appropriate 3D transport path can be generated.

[0142] (C6) In (C1) above, as shown in Figures 26, 29, etc., the transport path search unit 114 searches for an extended path (extended movement path) 1808 that will not collide with obstacles 1802 when the drone (moving object) 1701 is transported from the transport start position S0 to the transport end position G0, and will also minimize the overall transport time, thereby generating a three-dimensional transport path.

[0143] By performing the search described above, it becomes possible to avoid collisions that take into account the positional errors of moving objects, even when searching for transport paths in cases where objects can move freely in three-dimensional space, such as with the drone 1701.

[0144] (C7) In (C1) above, as shown in Figures 1 and 13, the path of the moving object is the theoretically shortest transport path (fastest transport path) that moves in a two-dimensional plane at the maximum possible speed. That is, the path creation unit 110 creates the theoretically shortest transport path in the case of ID=0 in the path creation method 211 of Figure 13. Then, the transport path search unit 114 generates a three-dimensional transport path based on a second extended transport path obtained by adding the first width described above and a third width based on the difference between the fastest transport path and the actual path of the moving object to the theoretically shortest transport path.

[0145] Since the actual path deviates from the fastest possible travel path, the accuracy of the 3D transport path can be further improved by considering a third width in addition to the theoretically shortest transport path, as described above.

[0146] (C8) In (C4) or (C5) above, the system further includes an output result creation unit 115 that outputs transport distance information and transport position information based on the three-dimensional transport path generated by the transport path search unit 114, as shown in Figures 9 and 12. The output transport distance information and transport position information may be numerical information such as the three-dimensional transport information 208 and three-dimensional transport distance information 209 shown in Figures 9 and 12, or they may be movement diagrams as shown in Figures 24 and 25. This transport distance information and transport position information can be used for transport control by the transport control device or for the operator to check the transport status.

[0147] (C9) As shown in Figures 1, 8, 21, 22, etc., the 3D transport path search method generates a 3D transport path in which the moving object and the transported object 1502 do not collide with obstacles 1505a to 1505c, based on an extended movement path obtained by adding a width based on the position error of the moving object (for example, the width (left) and width (right) described in the extended transport path information 207 in Figure 8, and the width between the pair of lines 1512 in Figure 22) to the movement path (path 1511) of the moving object (trolley 1501) that transports the transported object 1502, which is stored in the database 102, and a 3D map (3D map information) 204 which includes information on the size of obstacles 1505a to 1505c.

[0148] As described above, by considering errors related to the moving object (position errors of the moving object) when generating a 3D transport path that avoids collisions, it is possible to search for a highly accurate 3D transport path.

[0149] The embodiments and modifications described above are merely examples, and the present invention is not limited to these, as long as the features of the invention are not impaired. Other embodiments conceivable within the scope of the technical idea of ​​the present invention are also included within the scope of the present invention. For example, the embodiments described above are explained in detail to make the present invention easier to understand, and the present invention is not necessarily limited to those having all the described configurations. Furthermore, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, a part of the configuration of one embodiment may be added to the configuration of another embodiment. Furthermore, a part of the configuration of each embodiment may be added, deleted, or replaced with other configurations.

[0150] Furthermore, each of the aforementioned configurations, functions, processing units, and processing means may be implemented in hardware, for example, by designing them as integrated circuits, or they may be implemented in software by having a processor interpret and execute programs that realize each function.

[0151] Information such as programs, tables, and files that implement each function can be stored in memory, hard disks, SSDs (Solid State Drives), or on recording media such as IC (Integrated Circuit) cards, SD cards, and DVDs (Digital Versatile Discs).

[0152] Furthermore, the control lines and information lines shown are those deemed necessary for explanation purposes and do not necessarily represent all control lines and information lines required for implementation. In reality, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0153] 100...3D transport path search device (path search device), 101...Data input / output unit, 102...Database, 103...Path search unit, 110...Path creation unit, 111...Error estimation unit, 112...3D map processing unit, 113...Extended path creation unit, 114...Transport path search unit, 115...Output result creation unit, 200...Moving object information, 201...Error information, 202...Transport path information, 203...3D map processing width, 204, 1710...3D map, 205...Processed 3D map, 206...Path width, 207...Extended transport path information, 208...3D transport information, 209, 152 1,1730…3D transport distance information, 210…3D transport route, 211…Route creation method, 212…Ascending / descending control pattern, 1501…Trolley, 1502,1702…Transported object, 1505a~1505c,1509a~1509c,1609a~1609c,1712,1722,1802…Obstacle, 1511,1727,1728…Route, 1604…Extended 2D route, 1700a,1700b…Moving object, 1701…Drone, 1808,1810…Extended route, G,G0,G1…Transport end position, S,S0,S1…Transport start position

Claims

1. A three-dimensional transport path search device for searching for a three-dimensional transport path of a moving object that transports an object, A storage unit that stores: moving object information including information on each direction in which the moving object can move, the maximum speed during acceleration and deceleration for each direction of the moving object, acceleration and deceleration, the mass of the moving object, and the height of the moving object from the ground at the start of transport; upward / downward control pattern information which may or may not describe a plurality of upward / downward control patterns that represent the upward and downward methods of the transported object; extended movement path information which represents an extended two-dimensional movement path on a two-dimensional plane obtained by adding a first width based on the position error of the moving object to the movement path of the moving object; and three-dimensional map information which includes information on the position and size of obstacles. The system includes a transport path search unit that generates a three-dimensional transport path in which the moving object and the transported object do not collide with obstacles, based on the moving object information, the upward / downward control pattern information, the extended movement path information, and the three-dimensional map information stored in the memory unit, The transport path search unit, If the aforementioned upward / downward control pattern information contains the aforementioned multiple upward / downward control patterns, the upward / downward control pattern that minimizes the overall transport time and avoids collisions with obstacles interfering with the extended two-dimensional movement path when the transported object is transported along the extended two-dimensional movement path from the transport start position to the transport end position is selected from the aforementioned multiple upward / downward control patterns represented by the aforementioned upward / downward control pattern information to generate the three-dimensional transport path. A three-dimensional transport path search device that, if the plurality of upward and downward control patterns are not described in the upward and downward control pattern information, searches for an upward and downward control pattern that, when the transport of the moving object is carried along the extended two-dimensional transport path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional transport path, and minimizes the overall transport time, based on the moving object information, the extended transport path information, and the three-dimensional map information, and generates the three-dimensional transport path.

2. In the three-dimensional transport path search device according to Claim 1, The aforementioned moving object is a trolley or drone for an overhead crane. If the moving object is the trolley, the ascending / descending control pattern information contains the plurality of ascending / descending control patterns. A three-dimensional transport path search device in which, when the moving object is the drone, the ascending / descending control pattern information does not contain the plurality of ascending / descending control patterns.

3. In the three-dimensional transport path search device according to claim 1 or 2, The three-dimensional transport path search device calculates the first width based on at least the horizontal position error of the movement path of the moving object, by converting the horizontal position error into a horizontal width with the movement path as the central axis.

4. In the three-dimensional transport path search device according to claim 1 or 2, The system further comprises a 3D map processing unit that assigns a second width to the location and size information of the obstacles included in the 3D map information, based on at least one of the errors that occur when creating the 3D map information, the amplitude of the swing of the transported object, and the position error in the direction of travel when the moving object moves along the extended 2D movement path, and generates processed 3D map information. The transport path search unit is a three-dimensional transport path search device that generates the three-dimensional transport path based on the processed three-dimensional map information instead of the three-dimensional map information.

5. In the three-dimensional transport path search device according to claim 1 or 2, The movement path of the moving object is the fastest possible movement path that moves in a two-dimensional plane at the maximum possible speed, The transport path search unit generates the three-dimensional transport path based on a second extended movement path obtained by adding a first width and a third width based on the difference between the fastest movement path and the actual movement path of the object being transported to the fastest movement path.

6. In the three-dimensional transport path search device according to claim 1 or 2, A three-dimensional transport path search device further comprising an output result creation unit that outputs transport distance information and transport position information based on the three-dimensional transport path generated by the transport path search unit.

7. A method for searching for a three-dimensional transport path of a moving object that transports an object, using a computer and a database, The database stores information about a moving object, including information about the directions in which the moving object can move, the maximum speed during acceleration and deceleration for each direction of the moving object, acceleration and deceleration, the mass of the moving object, and the height of the moving object from the ground at the start of transport; information about an upward and downward control pattern, which may or may not describe a plurality of upward and downward control patterns representing the methods of upward and downward movement of the transported object; extended movement path information, which represents an extended two-dimensional movement path on a two-dimensional plane obtained by adding a width based on the position error of the moving object to the movement path of the moving object; and three-dimensional map information, which includes information about the position and size of obstacles. If the aforementioned upward / downward control pattern information contains the aforementioned plurality of upward / downward control patterns, the computer selects from the plurality of upward / downward control patterns represented by the upward / downward control pattern information the upward / downward control pattern that, when the moving object is transported along the extended two-dimensional movement path from the transport start position to the transport end position, will not collide with any obstacles interfering with the extended two-dimensional movement path and will result in the shortest overall transport time, thereby generating the three-dimensional transport path. A three-dimensional transport path search method, wherein if the plurality of upward and downward control patterns are not described in the upward and downward control pattern information, the computer searches for an upward and downward control pattern that, when the transport of the moving object along the extended two-dimensional transport path from the transport start position to the transport end position, does not collide with any obstacles interfering with the extended two-dimensional transport path, and the overall transport time is minimized, based on the moving object information, the extended transport path information, and the three-dimensional map information, and generates the three-dimensional transport path.

8. In the three-dimensional transport path search method according to Claim 7, The aforementioned moving object is a trolley or drone for an overhead crane. If the moving object is the trolley, the ascending / descending control pattern information contains the plurality of ascending / descending control patterns. A three-dimensional transport path search method wherein, when the moving object is the drone, the ascending / descending control pattern information does not contain the plurality of ascending / descending control patterns.

Images