Route generation device, route generation method, and route generation program

Abstract

To generate a more improved correction route.SOLUTION: A route generation device 10 includes: a map reading unit 1 for reading data of a map; a route search unit 2 for searching for a route by using a place of start and a destination in a designated map; and a route shaping unit 3 for applying an inspection pattern which include a route point and can be set to have different widths on the right and on the left with reference to the direction of travel of a route to a route point forming the route searched for, thereby setting the distance between the route point and an obstacle in the map.SELECTED DRAWING: Figure 12A

Description

[Technical Field]

[0001] The present invention relates to a route generation device, a route generation method, and a route generation program. [Background technology]

[0002] For example, Patent Document 1 describes a technology that searches for a route that a user can take based on information about a floor plan of a house, and then performs post-processing such as smoothing on the searched route to generate a final route. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6304771 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in conventional technology, when a path for an autonomous mobile robot or the like is generated using an algorithm such as RRT* (RRTstar), and then post-processing is performed to thin or smooth the path to generate the final path, the generated corrected path may intersect with obstacles such as walls.

[0005] An object of the present invention is to provide a route generation device, a route generation method, and a route generation program that are capable of generating an improved corrected route. [Means for solving the problem]

[0006] Among the inventions disclosed in this application, the outline of representative inventions will be briefly explained as follows.

[0007] A route generation device according to a representative embodiment of the present invention comprises: a map reading unit that reads map data; a route search unit that searches for a route using a designated departure point and destination point within the map; a setting unit that sets a distance between a route point that constitutes the searched route and an obstacle in the map by applying an inspection pattern that includes the route point and can be set to different widths on the left and right sides with respect to the traveling direction of the route; Equipped with. [Effects of the Invention]

[0008] The effects obtained by the representative inventions disclosed in this application will be briefly explained as follows.

[0009] That is, exemplary embodiments of the present invention can generate improved corrected paths. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 2 is a functional block diagram showing a control configuration of the path generation device according to the embodiment of the present invention. [Figure 2] 10A and 10B are diagrams illustrating an original route searched by a route search unit and a route shaped by a route shaping unit. [Figure 3] 3 is a partially enlarged view illustrating an example in which the corrected path CR interferes with an obstacle Mb in FIG. 2(2). [Figure 4] FIG. 10 is a partially enlarged view illustrating a process of providing a midpoint MP between adjacent route points CRn and CRn-1 that constitute a corrected route CR. [Figure 5] 10 is a partially enlarged view illustrating a travel direction vector Vdm of a corrected path CR, a right direction vector Vrd and a left direction vector Vld based on the travel direction vector Vdm. FIG. [Figure 6] 10 is a partial enlarged view for explaining the process of arranging the check pattern EP. FIG. [Figure 7] 10 is a partially enlarged view illustrating an example of interference between the inspection pattern EP and an obstacle Mb when the inspection pattern EP is moved. FIG. [Figure 8]Partial enlarged view illustrating the interference state with the obstacle Mb when moving the inspection pattern EP. [Figure 9] Partial enlarged view explaining the process of reconnecting to generate a path passing through the relay point. [Figure 10] Diagram comparing the path of the comparative example applying the prior art with the path created using the path generation device of the present embodiment. [Figure 11] Flowchart showing the form of the processing procedure in the path generation device shown in FIG. 1. [Figure 12A] Diagram explaining the relationship between the AMR to be used and the passage width and the inspection pattern according to Embodiment 2. [Figure 12B] Functional block diagram showing the control configuration of the path generation device in Embodiment 2. [Figure 13] Partial enlarged view explaining the states before and after the movement of the path when the inspection pattern according to Embodiment 2 is set with the parameter b and the parameter c having the same value. [Figure 14] Partial enlarged view explaining the process when the inspection pattern of Embodiment 2 is applied when the passage width W < a. [Figure 15] Partial enlarged view explaining the process when the inspection pattern of Embodiment 2 is applied when the passage width W ≥ a and there is no interference on the left side in the traveling direction. [Figure 16] Partial enlarged view showing the passage portion without interference on both the left and right sides in the traveling direction. [Figure 17] Diagram explaining an example of the process of reducing the width of the passage in the environmental map. [Figure 18] Diagram explaining the index for determining the parameters of the inspection pattern according to Embodiment 2. [Figure 19] Flowchart explaining the flow of the process when using the inspection pattern of Embodiment 2.

Mode for Carrying Out the Invention

[0011] <m (Embodiment 1) Hereinafter, embodiments of the present invention will be described with reference to the drawings. Each embodiment described below is an example for realizing the present invention, and does not limit the technical scope of the present invention. In the examples, components having the same function are given the same reference numerals, and repeated explanations thereof will be omitted unless particularly necessary.

[0012] <Summary> The path generation device described below is a device that uses an arbitrary environmental map to search for, shape, modify, set, etc. a path within the environmental map that an autonomous mobile robot or the like can travel. The path generation device of this embodiment can generate a modified path that is improved over conventional devices.

[0013] <Configuration of the route generation device> FIG. 1 is a configuration diagram showing an example of the configuration of a route generation device according to the first embodiment of the present invention. As shown in FIG. 1, the route generation device 10 includes an environment map information reading unit 1, a route search unit 2, a route shaping unit 3, a route correction unit 4, a route smoothing unit 5, and a storage unit 6. Although not shown, the route generation device 10 also includes a display unit such as a liquid crystal display, and an operation input unit such as a mouse or keyboard. Such a route generation device 10 can be implemented in a single computer, and can also include any peripheral devices such as a printer or a communication device.

[0014] Of the above, the blocks denoted by reference numerals 1 to 5 can be realized as functions carried out by a hardware processor such as a CPU or MPU. Separate processors may be used for each of these blocks as appropriate. The storage unit 6 may be any data storage medium such as a HDD, ROM, or RAM.

[0015] The environmental map information reading unit 1 performs a process of reading environmental map data from the storage unit 6. The type of environmental map is not particularly limited and can be any type, but unless otherwise specified, the following description will be based on the assumption that the environmental map is a map with paths formed that are wide enough for an autonomous mobile robot or the like to move through. In addition, the following description will be based on the assumption that any parts of the environmental map other than the paths are obstacles (see the black parts in Figure 2 as appropriate).

[0016] The environmental map data read by the environmental map information reading unit 1 is expanded into a working area such as RAM, and can then be displayed on the display unit described above. The read environmental map data can also be printed on paper using a printer or the like. Overall, the environmental map information reading unit 1 performs processing to read and output information on the environmental map before route setting.

[0017] The route search unit 2 searches for a route for an autonomous mobile robot or the like to travel from the output environmental map using an algorithm such as the RRT* (RRTstar) described above. Prior to the route search by the route search unit 2, a departure point (start point), an arrival point (goal point), appropriate intermediate waypoints, etc. can be set, for example, using the operation input unit described above. In this case, the route search unit 2 searches for a route that passes through the set departure point, arrival point, and intermediate waypoints from the output environmental map, and outputs the route resulting from the search. For ease of explanation, etc., the route searched by the route search unit 2 will sometimes be referred to as the "original route" below.

[0018] (1) at the top of Figure 2 shows the state in which the route search unit 2 searches for a route on the environmental map M output on the display (display screen) and outputs it as an original route OR. Note that the positions indicated by "START" and "GOAL" in the figure are the preset starting point (start point) and destination point (finish point), and this is also true for other figures. In the following, unless otherwise specified, it is assumed that no intermediate waypoints are set.

[0019] The route shaping unit 3 shapes and outputs the route of the output search result, i.e., the original route OR. Specifically, the route shaping unit 3 thins out route points of the original route OR output from the route search unit 2, moves the route points after the thinning out as appropriate within a range that does not cause them to come into contact with obstacles, connects adjacent route points after the thinning out, and outputs the result as a shaped route CR. Note that the shaped route CR output from the route shaping unit 3 can also be called a corrected route in the sense that it is a form in which the original route OR has been corrected, but it will be referred to as a shaped route here to distinguish it from the corrected route output from the route correction unit 4.

[0020] (2) at the bottom of FIG. 2 shows the state in which the original route OR has been reshaped by the route shaping unit 3 and output as a reshaped route CR on the screen of FIG. 2(1).

[0021] As a specific example of the display mode on the display screen, in (1) of Fig. 2, a layer (second layer) of the output original route OR is superimposed on a layer (first layer) of the environmental map M. In addition, in (2) at the bottom of Fig. 2, a layer of the reshaped route CR, which is reshaped by thinning out the route points of the original route OR, is superimposed as a third layer.

[0022] The route correction unit 4 further corrects the reshaped route CR reshaped by the route shaping unit 3 and outputs it as a corrected route (see the route indicated by symbol Rrj in FIG. 9 as appropriate). Specifically, the route correction unit 4 executes a series of processes including calculating required relay points, setting temporary relay points, setting a check pattern, calculating the movement direction of the check pattern, moving the check pattern, detecting intersections, determining relay points, and connecting route points. The right side of FIG. 1 shows the above processes in the route correction unit 4 as functional blocks.

[0023] That is, the route correction unit 4 has a relay point necessary location calculation unit 41, a temporary relay point setting unit 42, a check pattern setting unit 43, a check pattern movement direction calculation unit 44, a check pattern movement unit 45, an intersection determination unit 46, a relay point determination unit 47, and a route point connection unit 48. The functions of each of these units will be described later.

[0024] The route smoothing unit 5 performs a smoothing process on the route corrected (so to speak, re-corrected) by the route correcting unit 4, thereby smoothing the connections between the routes.

[0025] Next, with reference to FIG. 2(2) to FIG. 10, the problems in the prior art and the details of the processing executed by the route corrector 4 will be described in accordance with a specific example.

[0026] The enlarged portion shown on the right side of Figure 2(2) is a diagram illustrating the problems with the conventional route generation process. As shown in this enlarged portion, the original route OR output by the route search unit 2 does not interfere with the obstacle Mb, whereas the route of the reshaped route CR output by the route reshaping unit 3, which connects the route points CRn and CRn+1, intersects with, or interferes with, the obstacle Mb (in this example, the part where the passage turns at a nearly right angle).

[0027] It was found that this is because, in the process performed by the route shaping unit 3, the route points after thinning (in this example, at least one of route points CRn and CRn+1) are moved appropriately within a range that does not cause them to come into contact with the obstacle Mb, and these route points are moved so that the route connecting route points CRn and CRn+1 is as short as possible.

[0028] From another perspective, it has been found that by performing the above-mentioned processing in the route shaping unit 3, the distance between adjacent route points becomes longer at points where the route of the shaped route CR intersects with an obstacle Mb compared to points where the route does not intersect (in this example, the parts where the passage is straight) (see each route point of the shaped route CR shown in Figure 2(2) as appropriate).

[0029] Therefore, in this embodiment, the route correction unit 4 of the route generation device 10 performs the following process to reconstruct the route at the point where the route of the adjusted route CR intersects with the obstacle Mb.

[0030] First, the route correction unit 4 performs the function of the relay point required location calculation unit 41. Here, the relay point required location calculation unit 41 extracts all locations on the reshaped route CR output by the route reshaping unit 3 where the distance between route points exceeds a threshold, that is, all two route points CRn and CRn+1, among two adjacent route points on the reshaped route CR, whose distance between the points exceeds a threshold. More specifically, the relay point required location calculation unit 41 calculates the distance between two adjacent route points on the reshaped route CR, determines whether the calculated value exceeds the threshold, and extracts the two route points CRn and CRn+1 determined to exceed the threshold, repeating this process from the START route point to the GOAL route point on the reshaped route CR.

[0031] Here, the threshold value is not particularly limited, and can be arbitrarily set (initial value changed) by the user via the operation input unit described above. However, since the threshold value needs to be a value longer than the distances of the other route points that make up the reshaped route CR, an example of the initial value is set to twice the distances of the other route points.

[0032] Referring to (2) in Figure 2, in the reshaped route CR from START to GOAL, there are obstacles, in this example, wall sections that jut out to make right-angle turns, at four locations: the lower left, lower, upper, and upper right. Of these, the obstacle at the lower left does not intersect with either the original route OR or the reshaped route CR due to the position of the START point, and therefore the two corresponding route points on the reshaped route CR do not exceed the threshold. In contrast, the three obstacles at the lower, upper, and upper right do not intersect on the original route OR, but do intersect on the reshaped route CR.

[0033] In order to avoid repetition of explanation, the following will explain in detail the intersection with the upper obstacle Mb shown in the enlarged extracted view of Figure 2(2) and the processing performed by the path correction unit 4 to resolve such intersection, with reference to Figures 3 and beyond.

[0034] Figure 3 is a partially enlarged view illustrating the case where the reshaped route CR interferes with an obstacle Mb in Figure 2 (2). In Figure 3, arrow D indicates the direction of travel on the original route OR. Route points CRn and CRn+1 are two adjacent route points on the reshaped route CR whose distance exceeds a threshold, and the subscript n indicates the nth point when the route point at the START point is 1. The subscript n+1 indicates the next point after the nth route point, i.e., the next point in the direction of travel.

[0035] Next, the route correction unit 4 performs the function of the temporary relay point setting unit 42. Here, the temporary relay point setting unit 42 sets a midpoint MP as a temporary relay point between the route point CRn and the route point CRn+1 extracted by the relay point necessary location calculation unit 41.

[0036] 4 is a partially enlarged view illustrating the process of setting a midpoint MP (temporary relay point) between adjacent route points CRn and CRn+1 that make up the reshaped route CR. In this example, the midpoint MP is set at the midpoint between route points CRn and CRn+1 (see also FIG. 5(1)), but depending on the shape of the obstacle, the midpoint MP may be placed closer to route point CRn or, conversely, closer to route point CRn+1.

[0037] The temporary relay point setting unit 42 also calculates a traveling direction vector Vdm between the route point CRn and the route point CRn+1 extracted by the relay point required location calculation unit 41. Furthermore, the temporary relay point setting unit 42 defines the right and left sides of the calculated traveling direction vector Vdm. Here, Fig. 5 is a partially enlarged view illustrating the traveling direction vector Vdm of the reshaped route CR, and the right direction vector Vrd and left direction vector Vld based on the traveling direction vector Vdm.

[0038] That is, as can be seen by referring to the arrow D described above in Fig. 3 and (1) in Fig. 5, in this example, the travel direction vector Vdm is an arrow pointing from the route point CRn to the route point CRn+1, more specifically, an arrow pointing from the route point CRn to the route point CRn+1 via the midpoint MP. Also, referring to (2) in Fig. 5, in this example, an arrow obtained by rotating the travel direction vector Vdm 90 degrees clockwise is defined as the right direction vector Vrd, and an arrow obtained by rotating the travel direction vector Vdm 90 degrees counterclockwise is defined as the left direction vector Vld.

[0039] The definitions of the rightward direction vector Vrd and the leftward direction vector Vld are not necessarily limited to those described above. For example, the rightward direction vector Vrd may be defined as an arrow obtained by rotating the moving direction vector Vdm clockwise at any angle, such as 30 degrees, 45 degrees, or 60 degrees, and the same applies to the definition of the leftward direction vector Vld.

[0040] Next, the route correction unit 4 performs the function of the check pattern setting unit 43. Here, the check pattern setting unit 43 sets a check pattern EP as shown in (1) of FIG. 6 for the midpoint MP set by the temporary relay point setting unit 42. More specifically, the check pattern setting unit 43 arranges a circular check pattern (binary bitmap) with a radius b centered on the coordinates of the midpoint MP. Note that the value of the radius b is not particularly limited, and can be arbitrarily set (the initial value can be changed) by the user via the operation input unit described above.

[0041] Fig. 6(2) shows a state in which the check pattern EP shown in Fig. 6(1) is placed (set) at the midpoint MP. In the example shown in Fig. 6(2), the central part of the check pattern EP, i.e., the pattern center EPm, is aligned with the midpoint MP, and the midpoint MP and the check pattern EP are combined or integrated.

[0042] 6 illustrates a case where the outer edge shape of the check pattern EP is circular, but the shape is not limited to this and may be other polygonal shapes such as a rectangle, a hexagon, etc. As yet another example, the outer edge shape of the check pattern EP may be asymmetrical on the left and right, for example, a shape that combines semicircles with different diameters in the directions of the rightward vector Vrd and the leftward vector Vld.

[0043] Next, the path correction unit 4 performs the function of the intersection determination unit 46. Here, the intersection determination unit 46 determines whether the check pattern EP set by the check pattern setting unit 43 intersects with the corresponding obstacle Mb. In this example, the environmental map M for path generation is represented by a binary bitmap, and specifically, passable areas that serve as passages are represented by 1, and areas of obstacles Mb such as walls are represented by 0. Therefore, as a specific example of the intersection determination performed by the intersection determination unit 46, if any of the two-dimensional coordinate values ​​representing the outer edge of the circle of the check pattern EP matches the two-dimensional coordinate values ​​of the obstacle Mb in the environmental map M, i.e., represented by 0, then the intersection determination unit 46 determines that the check pattern EP does not intersect if any of the two-dimensional coordinate values ​​representing the outer edge of the circle of the check pattern EP does not match the two-dimensional coordinate values ​​of the obstacle Mb in the environmental map M, i.e., represented by 0.

[0044] Here, when the intersection determination unit 46 outputs a determination result that an intersection occurs, the path correction unit 4 performs the function of the check pattern movement direction calculation unit 44. Here, the check pattern movement direction calculation unit 44 performs a process of determining a direction in which to move the midpoint MP (temporary relay point) according to the determination result that an intersection occurs. As a specific example of the process by the check pattern movement direction calculation unit 44 of determining the direction in which to move the midpoint MP (temporary relay point), a determination is made to move the midpoint MP to either the right direction vector Vrd or the left direction vector Vld described above.

[0045] In one specific example, when the output result indicates that the obstacle Mb intersects with the obstacle Mb, the check pattern movement direction calculation unit 44 repeats the above determination process continuously up to a predetermined upper limit of the movement amount or up to a predetermined upper limit of the number of times (m times). The reason for this is that even if the direction in which the midpoint MP (temporary relay point) is moved (whether the rightward vector Vrd or the leftward vector Vld) is appropriate, the intersection with the obstacle Mb may not be resolved by moving the midpoint MP only once.

[0046] The above-mentioned amount of movement or the value of m can be arbitrarily set (initial value can be changed) by the user via the operation input unit. In one specific example, the upper limit (initial value) of the amount of movement is set to an integer multiple of the radius of the check pattern EP, for example, 2b (=b×2).

[0047] Next, the path correction unit 4 performs the function of the check pattern movement unit 45. Here, the check pattern movement unit 45 performs processing to move the midpoint MP (temporary relay point) in accordance with the movement direction (rightward vector Vrd or leftward vector Vld) determined by the check pattern movement direction calculation unit 44.

[0048] Then, the path correction unit 4 repeats the above-mentioned processes of intersection determination by the intersection determination unit 46, movement direction determination by the check pattern movement direction calculation unit 44, and movement of the check pattern EP and therefore the midpoint MP (temporary relay point) by the check pattern movement unit 45, until the outer edge of the check pattern EP no longer intersects with the obstacle Mb, up to the upper limit of the movement amount or the upper limit number of times.

[0049] In general, the check pattern moving unit 45 has the function of moving the (coordinate) position of the check pattern EP so that the check pattern does not interfere with the obstacle.

[0050] Thus, by carrying out the above-described series of processes, it is possible to eliminate interference between the obstacle Mb and the rectified route CR.

[0051] Here, when the above-mentioned intersection determination outputs a determination result indicating no intersection, the route correction unit 4 sequentially executes the functions of the relay point determination unit 47 and the route point connection unit 48. Here, the relay point determination unit 47 performs processing to set (confirm) the position of the check pattern EP and, in turn, the midpoint MP as a temporary relay point as a relay point. Then, the route point connection unit 48 performs processing to connect the route point CRn with the set (confirmed) route point, and further connect the route point with the next route point CRn+1.

[0052] On the other hand, due to various factors such as the complex shape of the obstacle Mb, there may be cases where the intersection with the obstacle Mb is not resolved even after repeating the above process. Figure 7 (1) and (2) are partially enlarged views illustrating an example of interference with the obstacle Mb when the check pattern EP is moved according to the rightward vector Vrd. In the example shown in Figure 7, it can be seen that the movement direction of the check pattern EP is invalid. In such cases, the path correction unit 4 performs the following process.

[0053] The path correction unit 4 performs the function of the check pattern movement unit 45. At this time, the check pattern movement unit 45 performs processing to move the midpoint MP (temporary relay point) in the direction opposite to the movement direction (rightward vector Vrd or leftward vector Vld) previously determined by the check pattern movement direction calculation unit 44. Then, as described above, the above-described processes of intersection determination by the intersection determination unit 46, movement direction determination by the check pattern movement direction calculation unit 44, and movement of the check pattern EP and therefore the midpoint MP (temporary relay point) by the check pattern movement unit 45 are repeated up to a preset movement amount or number of movements until the outer edge of the check pattern EP no longer intersects with the obstacle Mb.

[0054] 8(1), (2), and (3) are partially enlarged views illustrating interference with an obstacle Mb when the check pattern EP is moved in the direction of the leftward vector Vld, which is the opposite direction to that in FIG. 7. Similarly, here, an upper limit on the amount of movement or the number of times the check pattern EP can be moved is set in advance (for example, the upper limit on the amount of movement is set to 2b), and the check pattern EP is moved within that limit. Then, when the check pattern no longer intersects with the obstacle Mb, the midpoint MP is moved to that point and used as a relay point.

[0055] If the intersection with the obstacle Mb is not resolved even after this reverse movement process, the path correction unit 4 returns to the process of setting the check pattern EP by the check pattern setting unit 43. At this time, the check pattern setting unit 43 changes the value of the radius b described above in Fig. 6(1), for example, arranges a circular check pattern (binary bitmap) with the radius b set to half the value of the immediately preceding state, and repeats the processes of intersection determination, movement direction determination, and movement of the check pattern EP and ultimately the midpoint MP (temporary relay point).

[0056] Then, when the determination result indicates that the intersection with the obstacle Mb has been resolved, the route correction unit 4 inserts (sets or confirms) the relay point EPm as a route point and reconnects each route point, thereby generating a corrected route that corrects the above-mentioned shaped route. Fig. 9 is a partially enlarged view illustrating the reconnection process to generate a route that passes through the relay point EPm. Here, spline interpolation can be used as a specific example of the connection of route points, i.e., the connection process performed by the above-mentioned route point connection unit 48.

[0057] According to this embodiment, which performs such a series of processes, it is possible to generate an improved route (corrected route) including a rewired route Rrj that eliminates intersection (interference) with the obstacle Mb, as shown in Figure 9.

[0058] Thus, in the first embodiment, a check pattern EP having an outline larger than that of a midpoint MP between path points CR and CRn+1 that constitute the reshaped path CR and that are beyond the threshold distance is set. In this embodiment, the set check pattern EP and midpoint MP are then moved in either the rightward direction vector Vrd or the leftward direction vector Vld relative to the traveling direction vector Vdm. When the check pattern EP is moved to a position where it does not interfere with the obstacle Mb, the center of the check pattern EP, and therefore the midpoint MP, is used as a relay point to reconnect the check pattern. The path generation device 10 of this embodiment, which performs this series of processes, can generate a modified path that approximates the original path OR and does not interfere with the obstacle Mb by thinning out the number of path points in the original path OR and adding a minimum number of relay points, as shown in FIG. 9 .

[0059] Fig. 10 is a diagram showing a comparison between a route of a comparative example to which the conventional technology is applied and a route created using the route generation device 10 of this embodiment. (1) on the left side of Fig. 10 shows a modified route using the conventional technology method, which corresponds to the above-mentioned shaped route CR. In contrast, (2) on the right side of Fig. 10 shows a modified route created using the route generation device 10 of this embodiment.

[0060] In the conventional technology shown in Figure 10 (1), a modified path is generated using a spline interpolation and constant sampling method, and it can be seen that there is a portion of the path in the area circled in the upper left that interferes with a wall that bends to the right. In contrast, in the example shown in Figure 10 (2), the above-mentioned relay point EPm is inserted as a path point, and then a modified path is generated using the same spline interpolation and constant sampling method as in the conventional technology. In this way, by generating a modified path using the method of this embodiment, the interference with the wall that bends to the right is eliminated in the path in the area circled in the upper left.

[0061] Next, the flow of processing performed by the route generating device 10 will be described with reference to the flowchart in Fig. 11. Note that this flowchart corresponds to processing after the environmental map information reading unit 1 described above reads out the environmental map and displays it on the display (display screen) (see, for example, the environmental map M shown in Fig. 2), and after the starting point (START) and the destination point (GOAL) have been set.

[0062] In step S1, the route generation device 10 performs a route generation process by executing the above-described route search unit 2. Through this process, an original route OR (see (1) in FIG. 2) is displayed superimposed on the environmental map M that is read out on the display (display screen).

[0063] In step S2, the route generation device 10 executes the route shaping unit 3 to generate a shaped route CR. By the processing of step S2, a shaped route CR in which route points of the original route OR (see (1) in FIG. 2) have been thinned out, etc., is displayed (see (2) in FIG. 2).

[0064] In step S3, the route generation device 10 executes the relay point required location calculation unit 41 of the route correction unit 4 to extract all locations on the reshaped route CR where the distance between route points exceeds a threshold, that is, all two route points CRn and CRn+1 among two adjacent route points on the reshaped route CR where the distance between the points exceeds a threshold.

[0065] In step S4, the route generation device 10 executes the temporary relay point setting unit 42 of the route correction unit 4 to set a midpoint MP between the two extracted route points CRn and CRn+1 as described above in Fig. 4. Furthermore, the route generation device 10 executes the check pattern setting unit 43 of the route correction unit 4 to set a check pattern EP as described above in Fig. 6 on the midpoint MP.

[0066] In step S5, the route correction unit 4 calculates the traveling direction vectors of the two extracted route points CRn and CRn+1 (see arrow Vdm in FIG. 5). Note that the order of execution of step S4 and step S5 may be reversed.

[0067] In step S6, the path generating device 10 determines whether or not the path passing through the newly set midpoint MP (temporary relay point) intersects with an obstacle by executing the intersection determination unit 46 of the path correcting unit 4. As described above, this intersection determination can be performed based on whether or not the outer edge of the check pattern EP coincides with the obstacle Mb.

[0068] If the path generation device 10 determines that the path intersects with an obstacle (step S6: YES), the process proceeds to step S7. On the other hand, if the path generation device 10 determines that the path does not intersect with an obstacle (step S6: NO), the process proceeds to step S8.

[0069] In step S7, the path generation device 10 moves the check pattern EP and the midpoint MP according to either the rightward vector Vrd or the leftward vector Vld, in this example, the rightward vector Vrd, by executing the check pattern movement direction calculation unit 44 and the check pattern movement unit 45 of the path correction unit 4. Here, the path generation device 10 sets the moved midpoint MP as a tentative relay point EPm (see (2) in FIGS. 6 and 7).

[0070] Thereafter, the path generation device 10 returns to the determination step of step S6 and determines whether or not the tentative relay point EPm and the path including it intersect with an obstacle using the method described above.

[0071] Then, similarly to the above, if the path generation device 10 determines that an obstacle will be intersected (step S6: YES), the process proceeds to step S7. In the following step S7, the path generation device 10 moves the relay point EP in accordance with the leftward vector Vld, which is the opposite of the previous direction. The path generation device 10 sets the moved relay point EP as a temporary or new relay point EP (see (2) in FIG. 8).

[0072] In this way, the path generation device 10 repeats the process of determining whether the new relay point EPm (and the path including it) intersects with an obstacle in step S6 and the process of moving the temporary relay point EP in step S7 until it determines that the new relay point EPm (and the path including it) does not intersect with an obstacle (step S6: NO).

[0073] Then, when the intersection determination unit 46 determines that the check pattern EP and the relay point EP do not intersect with the obstacle Mb (step S6: NO), the path generation device 10 proceeds to step S8.

[0074] In step S8 after determining that the route does not intersect with the obstacle Mb (step S6: NO), the route generation device 10 executes the relay point determination unit 47 and the route point connection unit 48 of the route correction unit 4 to determine the relay point EP, reconnect the route that passes through the relay point EP, and proceeds to step S9.

[0075] In step S9, the route generation device 10 executes the relay point required location calculation unit 41 of the route correction unit 4 to determine whether or not there are any locations on the reshaped route CR where the distance between route points exceeds the threshold (requiring relay points). If the route generation device 10 determines that there are no locations requiring relay points (step S9: NO), it returns the process to step S4 and repeatedly executes the processes from step S4 onwards. On the other hand, if the route generation device 10 determines that there are no locations requiring relay points (step S9: YES), it determines that the route has been corrected for the entire route from the start to the goal, and ends the series of processes.

[0076] (Embodiment 2) Next, Embodiment 2 will be described. In the following, descriptions of configurations overlapping with those of Embodiment 1 described above will be omitted as appropriate.

[0077] In Embodiment 1, the path is automatically generated without particularly considering the width of the autonomous mobile robot (AMR) to be used and the width of the passageway.

[0078] In contrast, in Embodiment 2, a inspection pattern of a size considering these widths is set, and a modified path that maintains a certain distance from the edge of the passageway (usually with obstacles such as walls) serving as a constraint condition will be automatically generated will be described.

[0079] FIG. 12A is a diagram for explaining the relationship between the AMR to be used and the passageway width and the inspection pattern according to Embodiment 2. In the example shown in FIG. 12A, the following rules (constraint conditions) are provided as an application plan of the passage rules for the search path generated using a known algorithm such as RRT* (RRTstar). The terms "right" and "left" used below are both based on the traveling direction.

[0080] Constraint condition 1: When the passageway width W is less than a (m), that is, W < a, a target path TR is set leaving b (m) from the left of the passageway.

[0081] Constraint condition 2: When the passageway width W is greater than or equal to a (m), that is, W ≥ a, it is set to satisfy the following 2-1 and 2-2.

[0082] 2-1. Leave c (m) or more on the right of the passageway, and 2-2. Minimize the modification from the original path (the generated search path).

[0083] However, let a = b + c.

[0084] In this embodiment, the position of any waypoint on the original route OR (see FIG. 2(1)) generated by executing the above-described route search unit 2 is moved by executing the route shaping unit 3 so as to satisfy the above-described constraints 1 and 2. By moving the positions of the route points so as to satisfy such constraints 1 and 2, a target route TR that can accommodate two vehicles (in this example, two AMRs with the same width) passing through one passage is automatically generated.

[0085] Fig. 12B is a block diagram for explaining the function of the route shaping unit 3 according to the embodiment 2. Note that blocks equivalent to those shown in Fig. 1 in the embodiment 1 are given the same reference numerals, and their explanation will be omitted as appropriate.

[0086] 12B, the path shaping unit 3 of the second embodiment includes a check pattern setting unit 33, a check pattern movement direction calculation unit 34, a check pattern movement unit 35, an intersection determination unit 36, and a via point determination unit 37. Each of these blocks has a function similar to that of the blocks denoted by reference numerals 43 to 47 described above in FIG.

[0087] More specifically, the check pattern setting unit 33 sets a check pattern EP as shown on the right side of Fig. 12A for the original route OR generated by the route search unit 2. The check pattern EP will be described in detail later. The intersection determination unit 36 ​​determines whether the check pattern EP set by the check pattern setting unit 33 intersects with the corresponding obstacle Mb.

[0088] The check pattern movement direction calculation unit 34 determines to move the check pattern EP when the intersection determination unit 36 ​​outputs a determination result indicating intersection. The check pattern movement unit 35 performs processing to move the check pattern EP (temporary relay point) in accordance with the movement direction (for example, rightward or leftward) determined by the check pattern movement direction calculation unit 34.

[0089] 12A again, in one specific example of the second embodiment, the parameter b is set so that one side of the passage, in this example the left side, is as wide as possible for a vehicle (AMR) to pass through. The parameter c is set so that the other side of the passage, in this example the right side, is wide enough for one more vehicle (AMR) to pass through. This takes into consideration the so-called left-hand traffic rule.

[0090] In the second embodiment, a check pattern EP is used that can be set asymmetrically, as shown in Fig. 12A. This check pattern EP includes a semicircle hc1 with a radius (i.e., width) b set on the left side of the traveling direction, and a semicircle hc2 with a radius (i.e., width) c set on the right side of the traveling direction. The boundary between the semicircles hc1 and hc2 is located on the target route TR.

[0091] Here, the value of the radius b (width on the left side) of the semicircle hc1 in the check pattern EP and the value of the radius c (width on the right side) of the semicircle hc2 can be arbitrarily specified (the initial value can be changed) and set by the user via the operation input unit described above. Therefore, the check pattern EP of the second embodiment can be set to a symmetrical circle, in other words, b = c. Figure 13 is a partially enlarged view illustrating the state of the check pattern EP according to the second embodiment before and after the path movement when the parameters b and c are set to the same value.

[0092] 13, the path and obstacle Mb in the environmental map are represented by a bitmap, and a check pattern EP is shown in which the radii (widths) of the semicircles hc1 and hc2 are set to b = c. In this case, the basic configuration (shape) of the check pattern EP is the same as in the first embodiment, so the trajectory of the path after movement is expected to be equivalent to that in the first embodiment.

[0093] FIG. 14 is a partial enlarged view for explaining the process when the inspection pattern of Embodiment 2 is applied when the passage width W < a. The white arrows in FIG. 14 indicate the transition from the state before the movement of the inspection pattern EP and the path (left figure) to the state after the movement (right figure). This point is the same for FIG. 15.

[0094] When the passage width W is less than a (m), the path shaping unit 3 determines, by the function of the above-described intersection determination unit 36, whether or not the left side (semicircle hc1) of the inspection pattern EP interferes with the passage. Here, when the path shaping unit 3 determines that the left side (semicircle hc1) of the inspection pattern EP does not interfere with the passage, the process ends without moving the inspection pattern EP.

[0095] On the other hand, when the path shaping unit 3 determines that the left side (semicircle hc1) of the inspection pattern EP interferes with the passage, the inspection pattern EP is moved to the right as shown in FIG. 14 by the function of the above-described inspection pattern moving unit 35 until the left semicircle hc1 of the inspection pattern EP no longer interferes with the obstacle Mb, and the waypoint movement is completed.

[0096] FIG. 15 is a partial enlarged view for explaining the process when the inspection pattern of Embodiment 2 is applied when the passage width W ≧ a and there is no interference on the left side in the traveling direction. In this case, the path shaping unit 3 determines, by the function of the above-described intersection determination unit 36, whether or not the right side of the inspection pattern EP interferes with the passage. Here, when the path shaping unit 3 determines that the right side of the inspection pattern EP does not interfere with the passage, the process ends without moving the inspection pattern EP.

[0097] On the other hand, when the path shaping unit 3 determines that the right side of the inspection pattern EP interferes with the passage, the inspection pattern EP is moved to the left as shown in FIG. 15 by the function of the inspection pattern moving unit 35 until the semicircle hc2 of the inspection pattern EP no longer interferes with the obstacle Mb, and the waypoint movement is completed.

[0098] 16 is a partially enlarged view showing a passage portion where there is no interference on either the left or right side in the traveling direction. In this case, the path shaping unit 3 determines that the check pattern EP does not interfere with the passage by the function of the above-mentioned intersection determination unit 36, and does not move the path point by the check pattern movement unit 35.

[0099] In this way, by using a configuration in which a route is corrected using an inspection pattern EP that allows the width to be set on both the left and right sides of the direction of travel, the criteria for determining the interference situation with an obstacle Mb and the guidelines for the direction of movement of the inspection pattern EP to resolve the interference are clear, thereby improving the overall efficiency of processing in route generation.

[0100] In the above description, it is assumed that both the left and right sides of the check pattern EP can be set, but this is not limited to this, and it may be possible to set either the left or right side.

[0101] Next, another example of a route search or a route correction will be described with reference to Fig. 17 and Fig. 18. Fig. 17 is a diagram illustrating a processing example for reducing the width of a passage in an environmental map M. In Fig. 17, (1) in the upper part shows a state in which the width of the passage is reduced (compressed) by expanding obstacles Mb (hereinafter also referred to as "walls" for convenience) on the left and right sides of the passage, and then a route search is performed using the route search unit 2 (here, RRT*(RRTstar)). Meanwhile, (2) in the lower part of Fig. 17 shows the result of the route search in (1) in the upper part of Fig. 17 displayed on the environmental map before the expansion of the walls. Fig. 18 is a diagram illustrating an index for determining parameters of a test pattern according to the second embodiment.

[0102] When applying the check pattern EP according to the second embodiment, when generating a searched route using the RRT* or the like, it is preferable to take into account the vehicle width Rw of the AMR and perform a process of expanding the walls of the route so that the width of the narrowest passage among the passages through which the AMR wishes to pass is halved. Note that the entity that performs this process is not particularly limited, and for example, the route search unit 2 may perform this process as preprocessing for the route search. As described above, by performing a process of compressing the minimum width of the passage through which the AMR wishes to pass by half, the route search unit 2 can easily exclude passages that cannot accommodate two AMRs from candidates for the original route OR or the revised route CR. In this case, the route search unit 2 performs a route search so as to exclude passages whose compressed passage width (the width of the left and right walls) is less than the vehicle width Rw from candidates for the original route OR.

[0103] Furthermore, by performing such wall expansion processing (compression processing of the passage width), when the inspection pattern EP is subsequently used to determine whether the AMR intersects with the wall and to move along a path, the AMR can be prevented from coming into contact with the wall, and further, from getting closer to the wall than necessary.

[0104] For example, consider a case where the actual width of the aisle is 2 m and the corresponding display size on the display screen is 100 pix. In this case, if the AMR vehicle width Rw is 0.5 m, the display size will be 25 pix. If the minimum aisle width (W in FIG. 18) on the desired route is 0.6 m, the display size on the screen will be 30 pix. Therefore, by performing wall expansion (aisle width reduction) processing, the aisle width, i.e., the width between the left and right walls, will be 15 pix in display size. Furthermore, the parameter values ​​for the width b of the semicircle hc1 and the width of the semicircle hc2 in the test pattern EP can be set as follows: b = 25 pix, c = 50 pix.

[0105] Fig. 19 is a flowchart illustrating the flow of processing when using the check pattern EP according to embodiment 2. Here, the flowchart shown in Fig. 19 is based on the premise that the environment map M is read by the environment map information reading unit 1 and displayed on the display (display screen).

[0106] That is, in step S21 after the environmental map M is read, the route generation device 10 executes the route search unit 2 using RRT* (RRTstar) or the like to generate an original route OR for the passage of the environmental map M. In the following step S22, the route generation device 10 executes the route shaping unit 3 to apply the constraint conditions and the check pattern EP using the various parameters described above, thereby generating a corrected route CR (the above-mentioned target route TR). Then, in step S23, the route generation device 10 executes the route correction unit 4 and the route smoothing unit 5 to perform post-processing to appropriately smooth the corrected route CR generated (connected) by the route point connection unit 48 of the route correction unit 4.

[0107] <Simulation example> The inventors implemented the configuration of the above-described second embodiment in an actual model and performed a simulation. The simulation conditions were as follows: the image size of the display was 2540 (pixels) wide x 1140 (pixels) high, the display size of the width b of the semicircle hc1 of the test pattern EP was set to b=25 pixels, and the display size of the width c of the semicircle hc2 was set to c=50 pixels.

[0108] As a result, it was confirmed that the route generated by executing the route shaping unit 3 maintains a distance of at least b (m) from the left side of the aisle, and maintains a distance of at least c (m) from the right side where the aisle width is wider than b + c (m), and requires less travel distance compared to routes generated without executing the route shaping unit 3.

[0109] As described above, the route generation device 10 of the second embodiment includes an environmental map information reading unit (map reading unit) 1 that reads data of the environmental map M, a route search unit 2 that searches for a route using a specified departure point and destination point in the environmental map M, and a route shaping unit 3 that sets the distance between the route point and an obstacle Mb in the environmental map M by applying, to the route points that make up the searched route, an inspection pattern EP that includes the route point and can be set to different widths on the left and right based on the traveling direction of the route. With this configuration, a more improved corrected route CR can be generated.

[0110] The configurations of the first and second embodiments described above can be combined as appropriate depending on the purpose, etc. As a specific example in this case, after the route search unit 2 executes the process for generating an original route OR for the read environmental map M, the route shaping unit 3 performs a process for extracting a portion of the route or obstacle Mb that bends at a predetermined threshold angle (e.g., 90 degrees) or more. Then, the route shaping unit 3 applies the configuration of the second embodiment to the route or obstacle Mb other than the extracted portion, and applies the configuration of the first embodiment to the route or obstacle Mb of the extracted portion.

[0111] Furthermore, in the above-described first and second embodiments, the route generation device 10 is exemplified as a device that generates a route for a moving body to travel, but the present invention is not limited to this and can be applied to various other route generation devices, such as routes for human travel as described in Patent Document 1.

[0112] Furthermore, the present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment.

[0113] Furthermore, some of the configurations of each embodiment may be added to, deleted from, or replaced with other configurations. Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be implemented in hardware, in part or in whole, by designing them as integrated circuits, for example. Furthermore, the above-described configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as programs, tables, and files that implement each function may be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0114] Furthermore, the route generation device according to the above-described embodiment can generate improved routes, improving the efficiency of route generation work and reducing unnecessary processes and costs, thereby reducing overall energy consumption. This makes it possible to contribute to the Sustainable Development Goals (SDGs) led by the United Nations, particularly to Goal 7 (Ensure access to affordable, reliable, sustainable and modern energy) and Goal 13 (Take urgent action to combat climate change and its impacts). [Explanation of symbols]

[0115] 10...route generation device, 1...environmental map information reading unit, 2...route search unit, 3...route shaping unit, 4...route correction unit, 5...route smoothing unit, 6...storage unit, 33...inspection pattern setting unit, 34...inspection pattern movement direction calculation unit, 35...inspection pattern movement unit, 36...intersection determination unit, 37...via point determination unit, 41...relay point required location calculation unit, 42...temporary relay point setting unit, 43...inspection pattern setting unit, 44...inspection pattern movement direction calculation unit, 45...inspection pattern movement unit, 46...intersection determination unit, 47...relay point determination unit, 4 8...route point connection section, M...environment map, Mb...obstacle, OR...original route, CR...corrected route, CRn, CRn-1...adjacent route points after correction, D...arrow indicating direction of travel, MP...midpoint, Vdm...traveling direction vector, Vrd...right direction vector, Vld...left direction vector, EP...test pattern, b...radius of test pattern, EPm...center of test pattern, Rrj...reconnected route, TR...target route, W...passage width, hc1...semicircle with radius b, hc2...semicircle with radius c, RW...AMR vehicle width.

Claims

1. a map reading unit that reads map data; a route search unit that searches for a route using a designated departure point and destination point within the map; a route shaping unit that sets a distance between a route point constituting the searched route and an obstacle in the map by applying an inspection pattern that includes the route point and can be set to have different widths on the left and right sides based on the traveling direction of the route; Equipped with Route generation device.

2. The inspection pattern can specify at least one of the widths on the left and right sides based on the traveling direction of the path. The route generation device according to claim 1 .

3. 3. The route generation device according to claim 1, a compression unit that compresses the width of a path in the map to half the width of a vehicle; the route search unit excludes from the search target any passage whose compressed width is less than the vehicle width; Route generation device.

4. A computer comprising: Read the map data, A route search is performed using the designated starting point and destination on the map; a test pattern that includes the route points constituting the searched route and that can be set to have different widths on the left and right sides based on the traveling direction of the route, thereby setting the distance between the route points and the obstacles in the map; Route generation method.

5. Computer, a map reading unit that reads map data; a route search unit that searches for a route using a designated departure point and destination point within the map; a route shaping unit that sets a distance between a route point constituting the searched route and an obstacle in the map by applying an inspection pattern that includes the route point and can be set to have different widths on the left and right sides based on the traveling direction of the route; A route generation program to make it function as a

Images