Information processing apparatus, information processing method, data structure, and program

Abstract

To provide a technique that contributes to generation of a movement route in units of lanes by specifying a boundary line for each lane on a road having a plurality of lanes.SOLUTION: An information-processing device (1B) includes a boundary-line determination unit (11) that detects boundary lines (35a to 35d) that define travel paths (41 to 44) from map information and determines the boundary lines by classifying the boundary lines into boundary lines having end portions and boundary lines not having end portions, and when the travel path has a plurality of lanes (411,412,421,422,431,432,441,442), the boundary-line determination unit (11) sets a lane separation line (410420430440) that separates the plurality of lanes between the plurality of lanes.SELECTED DRAWING: Figure 14

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, a data structure, and a program. [Background technology]

[0002] In order to construct network data of roads on which vehicles travel, there is known a technique for generating centerlines and intersections from boundary line data of the roads. For example, Patent Document 1 describes a technique for generating centerlines and intersections from boundary line data of the roads.

[0003] The technology disclosed in Patent Document 1 divides images into three types: a first type that constitutes the outside of the road, a second type that constitutes the area around the road intersection, and a third type that constitutes the inside of the road.The inside of the road is then filled in, and the width of the road is narrowed to define the center line of the road. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-073009 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the conventional technology described in Patent Document 1 ultimately represents roads as a single line, evaluates intersections as points where lines intersect, and does not retain information about the width of the road or intersection. Furthermore, in standardized roads such as those in urban areas envisioned in Patent Document 1, intersections and other road edges can be easily distinguished by signs, shapes, etc., but in non-standardized road networks such as those in farmland, orchards, construction sites, and mines, it is not possible to mechanically set intermediate points for travel routes. Therefore, it is difficult to mechanically generate combinations of travel routes. Furthermore, the technology disclosed in Patent Document 1 does not generate road networks for each lane on roads with multiple lanes.

[0006] An object of one aspect of the present invention is to provide a technology that contributes to generating travel routes for each lane by identifying boundary lines for each lane on a road having multiple lanes. [Means for solving the problem]

[0007] In order to solve the above problem, an information processing device according to one embodiment of the present invention includes a boundary line identification unit that detects boundary lines that define a roadway from map data and classifies and identifies the boundary lines as boundary lines with ends and boundary lines without ends, and when the roadway has multiple lanes, the boundary line identification unit sets lane dividing lines that separate the multiple lanes between the multiple lanes.

[0008] An information processing method according to one embodiment of the present invention includes the steps of one or more processors detecting boundary lines defining a roadway from map data, classifying and identifying the boundary lines as those with ends and those without ends, and, if the roadway has multiple lanes, setting lane dividing lines separating the multiple lanes between the multiple lanes.

[0009] A data structure according to one embodiment of the present invention includes center point data derived by a center point derivation unit using boundary line data of a roadway, the center point pair data including the center point ID as an identifier, generated by a center point pair generation unit using the center point data, and intersection data generated by an intersection identification unit using the center point pair data, the intersection data including the center point pair ID as an identifier.

[0010] The information processing device according to each aspect of the present invention may be realized by a computer. In this case, the control program of the information processing device that causes the computer to operate as each part (software element) of the information processing device to realize the information processing device on the computer, and the computer-readable recording medium on which the control program is recorded, also fall within the scope of the present invention. [Effects of the Invention]

[0011] According to one aspect of the present invention, it is possible to provide a technology that contributes to generating travel routes for each lane by identifying boundary lines for each lane on a road having multiple lanes. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to a first embodiment of the present invention. [Figure 2] 1 is a flowchart showing the flow of an information processing method according to the first embodiment. [Figure 3] FIG. 2 is a diagram illustrating types of road boundary lines that appear in map data. [Figure 4] FIG. 10 is a diagram showing how to determine the center line of a road. [Figure 5] FIG. 10 is a diagram illustrating a situation in which the combination of center point pairs changes at an intersection. [Figure 6] FIG. 10 is a block diagram showing the configuration of an information processing device according to a second embodiment of the present invention. [Figure 7] FIG. 1 is a conceptual diagram illustrating division of a road into uniform sections and non-uniform sections. [Figure 8]This is an example of a road network from a start point to a destination point, including waypoints indicated by open circles. [Figure 9] 10 is an example of a cost map in which costs are assigned according to positions in the width direction of a road. [Figure 10] FIG. 10 is a diagram showing how the travel route changes when the temporary cutoff line is disabled and when it is enabled. [Figure 11] FIG. 10 is a diagram showing the positions at which temporary cutoff lines are generated at intersections using dotted lines. [Figure 12] FIG. 1 is a diagram showing the concept of treating a single-lane four-way intersection as an integrated intersection. [Figure 13] FIG. 10 is a block diagram showing the configuration of an information processing device according to a third embodiment of the present invention. [Figure 14] 1 is a schematic diagram of a four-way intersection where multiple roads intersect. DETAILED DESCRIPTION OF THE INVENTION

[0013] [Embodiment 1] (Configuration of information processing device 1) An embodiment of the present invention will be described in detail below with reference to the drawings. Fig. 1 is a block diagram showing the configuration of an information processing device 1 according to this embodiment. The information processing device 1 is a device that identifies a road on which a vehicle will travel (hereinafter also referred to as a "traveling path") from map data, and searches for a suitable traveling route including the position in the width direction of the traveling path when the vehicle travels on the traveling path.

[0014] Technological development is underway to use automated driving technology to move and work on work vehicles such as construction vehicles and agricultural vehicles. When moving such work vehicles, it is preferable to automatically drive them, including determining the route along which they should move and their travel position in the width direction of the travel path. The information processing device 1 according to this embodiment is an information processing device for searching for a route along which the vehicle should travel, including the position along the width direction of the travel path.

[0015] As shown in the figure, the information processing device 1 includes a boundary line identification unit 11, a center point derivation unit 12, a center point pair generation unit 13, an intersection identification unit 14, a processor 15, and a memory 16. The boundary line identification unit 11, the center point derivation unit 12, the center point pair generation unit, the intersection identification unit 14, the processor 15, and the memory 16 are connected to each other via a bus (not shown) so that they can communicate information with each other.

[0016] The above-mentioned units of the information processing device 1 do not need to be arranged together as a single device, and one or more of the units may be arranged in a distributed manner. Also, one or more of the units may be arranged on a cloud. This also applies to the following embodiments.

[0017] The processor 15 controls the entire information processing device 1. The memory 16 may include a primary memory and a secondary memory, and for example, the primary memory stores the control programs for each unit described below. The processor 15 loads the control programs for each unit into the secondary memory and executes them to realize the functions of each unit described below.

[0018] (Border line identification part 11) The boundary line identification unit 11 detects boundary lines that define the travel path from the map data and classifies and identifies them as boundary lines with edges and boundary lines without edges. The map data may be acquired from a geographic information system such as a GIS (Geographic Information System). Alternatively, GIS data is often not available for orchards, farmland, construction sites, etc. In such cases, map data may be created based on aerial photographs, etc. The map data preferably includes elevation data, road type, pavement type, presence or absence of obstacles, traffic restrictions, etc. A map including elevation data is a three-dimensional map. If the map data does not include such information, the user may input this information as appropriate.

[0019] The boundary line identification unit 11 detects boundary lines that define the travel lane from map data. The boundary lines of the travel lane are two lines that define the boundary in the width direction of the travel lane. There are no limitations on the method for detecting the boundary lines, but if GIS data is used, they can be acquired as the boundary part of the travel lane. If an aerial photograph is used, for example, continuous edges can be extracted and detected by image processing.

[0020] Boundary lines of a travel lane are divided into two types: boundary lines with ends and boundary lines without ends. Because map data has a limited range, boundary lines end at the outer edge of the map data area. The point where a boundary line ends at the outer edge of the map data is an end. Furthermore, even if the travel lane is within the area of ​​the map data, the dead end of the boundary line is the end. A "dead end" is a point beyond which the road is not considered to be part of the travel lane, such as a parking lot, a construction site, or an entrance / exit to a farm field. In other words, an end in this embodiment is the end of a boundary line to be processed. On the other hand, a boundary line without ends is a boundary line that forms a closed area in the map data.

[0021] The types of boundary lines will be described with reference to FIG. 3. FIG. 3 is a diagram illustrating the types of boundary lines of the roadway 30 that appear in the map data 100. The positions at the outer edge of the area of ​​the map data 100 or within the area where the boundary line of the roadway 30 ends are shown as end points E1 to E10. The boundary line identification unit 11 identifies these end points E1 to E10. Next, the boundary line identification unit 11 assigns boundary line numbers to the boundary line from end point E1 to end point E2. In FIG. 3, the boundary line from end point E1 to end point E2 is assigned the symbol 31a. The boundary line 31a is an example of a boundary line having end points.

[0022] Specifically, the boundary line 31a is generated as a set of short vectors. The boundary line identification unit 11 divides one boundary line into vectors whose direction change amount is smaller than a predetermined threshold, and records each vector in the memory 16 in association with a boundary line number (e.g., 31a).

[0023] Similarly, the boundary line identification unit 11 generates a boundary line 31b from end E3 to end E4, a boundary line 31c from end E5 to end E6, a boundary line 31d from end E7 to end E8, and a boundary line 31e from end E9 to end E10. Note that end E4 and end E5 are dead-end ends of the travel path 30.

[0024] Furthermore, the boundary line identification unit 11 generates an end boundary line (hereinafter also referred to as an "end line") that connects the ends of two boundary lines having ends. For example, as shown in FIG. 3, the boundary line identification unit 11 generates an end line 35a by connecting an end E1 to an end E10 located immediately adjacent to it. Similarly, for other locations, the boundary line identification unit 11 connects the nearest ends to generate end lines 35b, 35c, 35d, and 35e. The boundary line identification unit 11 assigns an identification code to the generated end line and records it in the memory 16.

[0025] Edge lines are lines that cross the roadway at the outer edge of the map data, or are entrances and exits to fields or parking lots, etc., so when plotting the map data, the user can draw the roadway boundary line and edge lines separately, and the boundary line identification unit 11 can detect them based on the attribute information it holds. Alternatively, the boundary line identification unit 11 can draw the roadway boundary line including the edge lines, and then automatically extract them based on the length and shape of the boundary line.

[0026] When all boundary lines having ends and end lines are connected, one closed line is obtained. The boundary line identification unit 11 can assign unique identification codes and predetermined circumferential direction information to the boundary lines and end boundary lines of the area closed by the boundary lines having ends and end lines. The unique identification codes of the boundary lines and end boundary lines are, for example, the codes 31a to 31e and 35a to 35e assigned to each boundary line and end line. The predetermined circumferential direction information is, for example, information on the clockwise direction. In FIG. 3, a clockwise arrow is assigned to the line connecting all boundary lines having ends and end lines.

[0027] On the other hand, boundary lines 32a and 32b are boundary lines that have no ends. Boundary lines 32a and 32b are connected lines and each form a closed area. Boundary line identification unit 11 can assign a unique identification code to the boundary line of the area closed by the boundary line that has no ends, and circumferential direction information in the opposite direction to the above-mentioned predetermined circumferential direction information. For example, the unique identification codes to the boundary line that has no ends are codes 32a and 32b, and the circumferential direction information in the opposite direction to the predetermined circumferential direction information is, for example, information indicating a counterclockwise rotation.

[0028] The boundary line identification unit 11 divides the boundary line without ends into vector lines whose change in direction is smaller than a predetermined threshold, assigns a code associated with the boundary line number to each vector line, and records the code in the memory 16. In other words, the boundary line identification unit 11 assigns a counterclockwise direction to the vector lines constituting the boundary lines 32a and 32b without ends, which is opposite to the direction of the vector lines assigned to the boundary lines with ends. In Figure 3, the boundary lines 32a and 32b without ends are assigned counterclockwise arrows.

[0029] Through the above process, as shown in Fig. 3, the travel path 30 can be defined as an area sandwiched between a boundary line formed by clockwise vectors and a boundary line formed by counterclockwise vector lines. More specifically, the travel path 30 is defined as an area inside an area formed by connecting clockwise vector lines and an area outside an area formed by connecting counterclockwise vector lines. In this case, the boundary line identification unit 11 identifies the area to the right of the clockwise vector line and the area to the right of the counterclockwise vector line as the travel path 30.

[0030] As described above, the specification of the travel path 30 is generally performed as follows: That is, the inclusion relationship of areas (polygons) closed by boundary lines with no end portions, or areas (polygons) closed by boundary lines with end portions and end lines, is specified, the circumferential direction of each area (polygon) is set alternately in the forward direction (for example, clockwise) and the reverse direction (for example, counterclockwise) in the order of inclusion, and the area on a predetermined side in the circumferential direction is specified as the travel path.

[0031] More specifically, the area of ​​the track 30 can be determined by grasping the inclusion relationships between polygons, regardless of whether they have edges. Both boundary lines with and without edges can be represented as polygons, including the edge lines. A polygon is a closed figure formed by connecting short straight lines. Polygons must not intersect with each other. A single polygon may contain one or an even number of polygons. In other words, the number of polygons contained within the boundary polygon of the outermost track 30, excluding the number of polygons within itself, is either zero or an odd number. When multiple polygons exist within the outermost polygon, the polygons are grouped into odd-numbered polygons and even-numbered polygons, counting from the outermost polygon. The odd-numbered polygons are designated as clockwise and the even-numbered polygons as counterclockwise. The right side of the circumference is the track area.

[0032] (Center point derivation unit 12) Next, a method for determining the center line of the travel path 30 will be described with reference to the drawings. 401 in FIG. 4 is an enlarged view of portion X in FIG. 3, illustrating how to determine the center line of the travel path 30. The center point derivation unit 12 derives a plurality of first center points, which are the centers of circles or spheres with a minimum radius that are tangent to a first boundary line that defines the travel path and inscribed in a second boundary line that is different from the first boundary line. Furthermore, the center point derivation unit 12 associates the first center points with the first boundary line and records them in the memory 16.

[0033] Specifically, as shown in 401 of FIG. 4, the center point derivation unit 12 derives the position of a center point CP31a1 of an inscribed circle C31a1 that is tangent to the starting point 31a1 of one vector constituting the boundary line 31a (first boundary line) and is tangent to a boundary line 32a (second boundary line) different from the boundary line 31a. The center point CP31a1 derived in this manner is also referred to as the first center point. The center point derivation unit 12 assigns a unique identification code to the derived center point CP31a1, associates the identification code with one of the boundary lines 31a, and records the association in the memory 16. The vectors (circumferential direction information) of the first boundary line and the second boundary line are in opposite directions.

[0034] Similarly, the center point derivation unit 12 derives the position of the center point CP31a2 of the inscribed circle C31a2, which is tangent to the start point 31a2 of the next vector and to a second boundary line 32a different from the first boundary line 31a. The center point derivation unit 12 assigns an identification code (center point ID) to this center point CP31a2, associates the identification code with the first boundary line 31a, and records the associated identification code in the memory 16. The center point derivation unit 12 records a vector connecting the center point CP31a1 and the center point CP31a2 in the memory 16 as a center line vector CPV31a1. Hereinafter, the center line vector will also be referred to as a center point column vector. In this way, the center point derivation unit 12 finds the first center point for each of the vectors constituting the boundary line 31a, and records the associated center point in the memory 16 in the memory 16. The center point derivation unit 12 performs the above process for the boundary lines 31a to 31e, which have end portions.

[0035] In addition, the center point derivation unit 12 may record in memory 16, as information regarding the first center point, information such as the center point identification code (center point ID, for example, 31a1) and boundary line number (for example, 31a), as well as the position of the center point CP31a1, the width of the road 30 at the center point position (diameter of the inscribed circle C31a1), the distance between the center point and the adjacent center point CP31a2, traffic restrictions (speed limit, one-way street, height limit, width limit, weight limit, etc.), altitude (or elevation difference from a specified position), and whether or not the road is paved, in association with the center point code.

[0036] Meanwhile, the center point derivation unit 12 derives the position of the center point CP32a1 of the inscribed circle C32a1, which is tangent to the starting point 32a1 of one vector constituting the boundary line 32a having no end portion and is tangent to a boundary line 31a different from the boundary line 32a. Therefore, in this case, the boundary line 32a is the first boundary line, and the boundary line 31a is the second boundary line. The center point derivation unit 12 associates the identification code of the derived center point CP32a1 with the boundary line 32a and records them in the memory 16.

[0037] Similarly, the center point derivation unit 12 derives a center point CP32a2 of an inscribed circle C32a2 that is tangent to the start point 32a2 of the next vector and to a second boundary line 31a different from the first boundary line 32a. The center point derivation unit 12 associates the identification code of this center point CP32a2 with the boundary line 32a and records them in the memory 16. The center point derivation unit 12 may also record information other than the identification code of the first center point and the boundary line number, as described above. The center point derivation unit 12 records a vector connecting the center points CP32a1 and CP32a2 in the memory 16 as a center point column vector CPV32a1.

[0038] In this way, the center point derivation unit 12 finds the center point for each of all vectors that make up the boundary line 32a, and records the center point in association with the boundary line 32a in the memory 16. The center point derivation unit 12 performs this process for all boundary lines 32a, 32b that do not have end portions.

[0039] As shown in 402 of FIG. 4, the center point derivation unit 12 generates a center point column vector (vector connecting center points) that sequentially connects center points CP31a1, CP31a2, ..., to generate center line CL31a. The center point derivation unit 12 also generates a center point column vector that sequentially connects center points CP32a1, CP32a2, ..., to generate center line CL32a. As shown in 402 of FIG. 4, the center line is a center point column vector. The center point derivation unit 12 records all of the information about the center point column vector included in center line CL31a in memory 16. The center point derivation unit 12 also records all of the information about the center point column vector included in center line CL32a in memory 16. Note that, although center line CL31a and center line CL32a are depicted separated from each other in 402 of FIG. 4, this is for illustrative purposes only; in reality, the two center lines coincide or nearly coincide.

[0040] Note that FIG. 4 illustrates the case where two-dimensional map data is used. However, three-dimensional map data may also be used. In this case, the boundary lines are also generated three-dimensionally. For example, if the travel path 30 has height restrictions, width restrictions, or obstacles in the width or height directions, the boundary line identification unit 11 generates multiple three-dimensional boundary lines to define the outer edges of a spatial region without such restrictions. Then, the center point derivation unit 12 derives the center point of an inscribed sphere with the smallest radius inscribed in the three-dimensional boundary lines.

[0041] (Center point pair generation unit 13) After the above process is completed, the center point pair generation unit 13 associates the first center point associated with the first boundary line with the second center point associated with a second boundary line different from the first boundary line and located closest to the first center point. A pair of a first center point and its closest second center point is referred to as a center point pair (or center point pair data). "Associated with a boundary line" is also referred to as "having a boundary line number." The center point pair is generated by searching for a center point (e.g., a second center point) that has a boundary line number other than the boundary line number of a certain center point (e.g., a first center point) nearest to the first center point and storing the identification code of the nearest center point as attribute information. In the case of a single track, two center lines are generated for the two boundary lines that make up the track 30. At points on the track 30 other than intersections, the center points of these two center lines form a pair. However, at intersections, the combination of center point pairs changes.

[0042] FIG. 5 is an enlarged view of portion Y in FIG. 3, illustrating how the combination of center point pairs changes at a three-way intersection. The black circles in FIG. 5 indicate center points (starting points of center point column vectors). As indicated by 501 in FIG. 5, center line CL31a runs along boundary line 31a. Center line CL31b runs along boundary line 31b. Center line CL32a runs along boundary line 32a. Note that within intersection Z in FIG. 5, center lines CL31a, CL31b, and CL32a are drawn far apart, but this is for ease of explanation; in reality, the three center lines are not far apart.

[0043] 502 in FIG. 5 is an enlarged view of intersection Z of 501. As shown in 502, for example, the center point of center line CL31a is close to the center point of center line CL32a in the upper right corner, and is therefore paired with the center point of center line CL32a. The pair of center points of center lines CL31a and CL32a changes to pair A2, pair A1, and pair A as one approaches the intersection. However, as one moves further into the intersection, the center point of center line CL31a pairs with the center point of center line CL31b, as shown in pair B.

[0044] Similarly, the pair of center points of center line CL31b changes from pair B2 to pair B1 to pair B as shown in the lower right before entering intersection Z. However, as it further enters intersection Z, the center point of center line CL31b pairs with the center point of center line CL32a as shown in pair C. Furthermore, as it enters intersection Z from the lower left, the pair of center points of center line CL32a changes from a pair with the center point of center line CL31b as shown in pair C to a pair with the center point of center line CL31a as shown in pair A.

[0045] (Intersection identification section 14) Using this property, the intersection identification unit 14 identifies the position of the first center point where the combination of two boundary lines associated with the first center point and the second center point changes as information indicating an intersection of a travel lane. Specifically, the intersection identification unit 14 assigns an identification code of the intersection Z (e.g., Z) to the center point where the pair changes at the intersection Z (the center point indicated by the double circle with a solid background in 502 in FIG. 5). In the case of a single track, the center points generated from the boundary lines facing each other across the travel lane 30 are extracted. The center points indicating the intersection are identified for each boundary line. For example, in a three-way intersection, there are three change points (the double circle in 502 in FIG. 5). In the case of an n-way intersection, there are 3(n-2) points. However, since the same point may be identified, the correct number is 3(n-2) or less. The intersection identification unit 14 assigns a center point pair ID to the center point pairs at the intersection and stores them.

[0046] For example, at a three-way intersection, three center points are identified as the intersection, so one intersection has information on three center points. Therefore, the representative coordinates of the intersection may be the average coordinates of the three points. Furthermore, when there are many intersecting roads or when the intersection is made up of multiple lanes, multiple intersections as described above will occur, so multiple intersections may be combined into one intersection, taking into account the distance between each intersection. This intersection is called an integrated intersection. Integrated intersections are associated with information on the center points that indicate the multiple intersections before they were combined. This information held by integrated intersections can be used when obtaining the adjacent structure of the intersection, which will be described later. Integrated intersections will be discussed later.

[0047] The center point identified as an intersection is associated with information on the center points to adjacent intersections. As described above, each center point may be associated with roadway information such as the width of the roadway 30, the distance between the center points (from which the distance between adjacent intersections can be calculated), traffic restrictions, elevation, obstacles, and the presence or absence of pavement. This information can be used when searching for a vehicle's driving route, including the position of the roadway in the width direction.

[0048] The information processing device 1 having the above configuration can set waypoints even on an irregular road network and identify the optimal driving route that passes through those waypoints. By using such technology, it becomes easier to use map data to find out which position on the road a special vehicle, such as a construction vehicle or an agricultural vehicle, should take.

[0049] [Embodiment 2] Other embodiments of the present invention will be described below. For ease of explanation, the same reference numerals will be used to designate components having the same functions as those described in the above embodiment, and the description thereof will not be repeated.

[0050] 6 is a block diagram showing the configuration of an information processing device 1A according to a second embodiment of the present invention. As shown in the figure, the information processing device 1A includes a boundary line identification unit 11, a center point derivation unit 12, a center point pair generation unit 13, an intersection identification unit 14, an edge acquisition unit 17, a section information recording unit 18, an adjacent information identification unit 19, a route setting unit 20, a route identification unit 21, a uniform section extraction unit 22, a via point setting unit 23, a deadline setting unit 24, a processor 15, and a memory 16. The above-mentioned units, the processor 15, and the memory 16 are connected to each other via a bus (not shown) so as to be able to communicate information with each other.

[0051] The functions of the boundary line identification unit 11, center point derivation unit 12, center point pair generation unit 13, intersection identification unit 14, processor 15, and memory 16 are the same as those described in embodiment 1, and therefore description thereof will be omitted. That is, the information processing device 1A includes an edge acquisition unit 17, a section information recording unit 18, an adjacent information identification unit 19, a route setting unit 20, a route identification unit 21, a uniform section extraction unit 22, a via point setting unit 23, and a deadline setting unit 24 in addition to the configuration of the information processing device 1 according to embodiment 1.

[0052] The edge acquisition unit 17 acquires the edge of a boundary line from map data. As described above, the edge is the edge of the boundary line to be processed. For example, the edge acquisition unit 17 can acquire, as the edge, a point where the boundary line detected by edge acquisition processing of the image ends. The edge can be one of the start point or destination point of the travel route.

[0053] A specific example of a method for detecting edges will be described below. For example, if the line segment vectors constituting the boundary line 31 change gradually before and after and the section has a constant length, the edge is usually about the width of the road, so it can be determined that it is not the edge of the boundary line. Conversely, if a line segment vector changes suddenly, and then the next line segment vector also changes suddenly (the angle that the next line segment vector forms with the first line segment vector is a right angle or an angle close to that), and the magnitude of the first line segment vector is about the width of the road, it can be determined that it is the edge of the boundary line.

[0054] (Adjacent information identification unit 19) The adjacent information identification unit 19 identifies adjacent information of two adjacent intersections by referring to the identification codes of the center points that indicate the intersections. The adjacent information of intersections is information that indicates which intersections are directly connected to each other. As described above, the center points that indicate intersections are given identification codes of the center points in addition to boundary line numbers, so the adjacent information of intersections can be identified by the order of the boundary line numbers and the identification codes of the center points. The adjacent information identification unit 19 identifies adjacent information of intersections by using section information recorded in memory 16. The adjacent information identification unit 19 may identify the width of the travel lane, the section distance, traffic restrictions, pavement type (e.g., asphalt pavement, earth pavement), road type (e.g., national highway, prefectural road, municipal road, private road), etc. as adjacent information.

[0055] (Section information recording unit 18) The section information recording unit 18 records at least one of information on the width of the road, the section distance, and traffic restrictions for the section between two adjacent intersections as section information for that section in the memory 16. As described above, the center point derivation unit 12 associates the code of the center point with the road information at that center point and records them in the memory 16. The section information recording unit 18 obtains from the memory 16 information on the road for the section between the two adjacent intersections that includes at least one of the width of the road, the section distance, and traffic restrictions from the road information associated with the code of the center point, and records this information as section information.

[0056] Next, a method for setting a vehicle's driving route will be described. When large, specialized vehicles such as construction vehicles (dump trucks, trucks, self-propelled cranes, etc.) or agricultural vehicles (combine harvesters, tractors, etc.) are driven automatically, the driving position in the width direction of the road is also important. For example, in areas adjacent to bends or intersections, the driving route must be determined taking into account the vehicle's turning radius (or wheelbase length). Therefore, in order to search for the optimal driving route in the width direction of the road, it is preferable to search separately for bends or intersections and other areas. This method will be described below.

[0057] FIG. 7 is a conceptual diagram illustrating how a road is divided into uniform sections and non-uniform sections. The uniform section extraction unit 22 extracts uniform sections of the road 30 based on the curvature of the road 30 of the vehicle. The curvature of the road 30 is an index indicating the degree of curvature of the road 30. For example, the amount of change in the direction of the center point column vector can be considered as the local curvature of the road 30. Therefore, the uniform section extraction unit 22 may extract uniform sections of the road 30 based on the rate of change of the center line vector, which indicates the local curvature of the road 30.

[0058] For example, the uniform section extraction unit 22 can extract a section where the change in the direction of the center point column vector is equal to or less than a predetermined threshold (a substantially straight road), or a section where the increment of the center point column vector is substantially constant (a curve with a constant curvature) as a uniform section. The uniform section extraction unit 22 extracts sections other than uniform sections as non-uniform sections. Specifically, as shown in Fig. 7, a range of a straight road or a curve with a constant curvature is extracted as a uniform section, and a range of an intersection or a bend is extracted as a non-uniform section.

[0059] The uniform section extraction unit 2 may also add conditions for extracting a uniform section, such as that the cost map of the travel path 30, which will be described later, is approximately symmetrical with respect to the center line position, or that the width is sufficiently larger than the width of the vehicle. Also, a condition may be added that if there is an obstacle, such as a utility pole, installed close to the travel path, the section will not be extracted as a uniform section.

[0060] The via point setting unit 23 also sets via points in the end regions of the uniform section. The end region of the uniform section is a region that includes the end of the uniform section and an area extending from the end to a certain extent inside the uniform section. The via points are boundary points (nodes) that limit the search range when searching for an optimal driving route. If the uniform section is adjacent to an intersection or a bend, some vehicles may start turning within the uniform section. Therefore, as shown in FIG. 7, the via point setting unit 23 sets via points at a distance that takes into account the vehicle's turning radius from the end of the uniform section. The larger the vehicle's turning radius, the greater the distance that the via point setting unit 23 sets via points at from the end of the uniform section. This distance can be determined, for example, based on the vehicle's turning radius (or wheelbase length).

[0061] In the uniform section, the positioning of the via point in the width direction of the road is performed by taking the uniform section to be of sufficient length, creating a corresponding cost map, and calculating the travel path of the moving body. Since the travel path will converge to a fixed position (approximately the position of the center line), the via point setting unit 23 sets the position of the center line at the end as the via point.

[0062] Generally, when setting a vehicle's travel route, the concepts of lanes and nodes are often used. For example, intersections are defined as nodes, routes connecting the intersections are defined as lanes, and a travel route is set by combining nodes and lanes. While this process is easy on standardized roads, setting nodes and lanes is difficult on orchards, farmland, construction sites, etc. However, in this embodiment, a vehicle's travel route can be easily set by setting waypoints as nodes.

[0063] The way point setting unit 23 may also set way points on edge line portions of the travel path. These edge line portions can be the start point or the destination point. The way point setting unit 23 may set the way point on the edge line to, for example, the midpoint of the edge line.

[0064] FIG. 8 shows an example of a road network from a start point S to a destination point G, including via points indicated by open circles. Of the sections between adjacent via points, solid lines indicate uniform sections, and dotted lines indicate non-uniform sections. The route setting unit 20 sets a combination of travel routes from the start point S (starting point) to the destination point G (ending point) that pass through the via points. A known method can be used as a method for setting all combinations of travel routes from the start point S to the destination point G.

[0065] For example, the route setting unit 20 searches for a center point adjacent to the above-mentioned via point and adds that center point to the intersection adjacency structure, thereby outputting all combinations of travel routes with the via point as the starting and ending point. Here, a travel route is a route that includes the position in the width direction of the travel path. Also, a travel route is a route that indicates which travel path a vehicle will take to travel to the destination, without including the position in the width direction of the travel path.

[0066] The route setting unit 20 also sets a travel route including a travel position in the width direction of the road between two adjacent waypoints on the road network shown in Fig. 8. The route setting unit 20 can set a travel route by, for example, combining a hybrid A* algorithm and an MPC method.

[0067] Specifically, based on a cost map generated from the boundary line and the center line of the road, the hybrid A* algorithm is used to generate a sequence of path points that will result in a low cost for the vehicle to evaluate. However, this algorithm generates a sequence of path points that does not guarantee continuity in the vehicle's yaw angle, so the changes in curvature are discontinuous. For this reason, a model predictive control (MPC) method, which can take into account the continuity of the curvature of the driving path, is used to perform route calculation again using the sequence of path points obtained by the hybrid A* algorithm as the target point sequence, and a sequence of path points with smooth curvature (a sequence of path points with continuously changing curvature values) is obtained.

[0068] As described above, the process of acquiring a sequence of route points may be repeated twice to generate a driving route. In other words, sections with large curves are identified based on the curvature obtained in the first acquisition process. Then, in the second acquisition process, a different evaluation point position is used to correct the driving route around the curve. Specifically, an evaluation point is set at a position away from the center of the vehicle in the direction of the turn, and the cost is evaluated. Because road width information is included in the center point sequence, the position of the evaluation point can be changed during route calculation. By changing the evaluation point in this way, the cost is reduced by traveling on the outside of the driving path around a curve. In other words, a driving route is generated that bulges outward on a curve rather than staying in the center of the driving path. This reduces the possibility of a large vehicle running off the road. When generating such a driving route, it is not necessary to use the Hybrid A* algorithm; various algorithms can be used.

[0069] Even for roads with multiple lanes, the above processing can be applied to intersections with multiple lanes by drawing the road lanes as polygons with boundary lines having the width of the road lanes.

[0070] If there are no via points, a single continuous driving route must be created from a certain departure point to the destination, which results in an enormous number of combinations of driving routes, making cost calculation time long and unrealistic. Therefore, by individually evaluating the driving routes between via points and connecting the driving routes between the via points, the number of combinations of driving routes can be reduced to a number that is realistically manageable.

[0071] Next, the route identification unit 21 refers to the cost map of the travel route, calculates the travel cost of the travel route set by the route setting unit 20, and identifies the optimal travel route. Specifically, the route identification unit 21 calculates the travel costs of multiple travel routes set by the route setting unit 20 for each of the uniform sections and non-uniform sections included in the extracted movement route. Then, the optimal travel route is identified by comparing the travel costs of the multiple travel routes. The optimal travel route is, for example, a route with the smallest travel cost. The travel cost is the total cost when a certain vehicle follows a certain travel route. The travel cost can be calculated, for example, using a cost map generated from boundary lines and the center line of the travel route.

[0072] Specifically, the route identification unit 21 calculates the travel cost by referring to a cost map in which costs are assigned according to the position in the width direction of the travel path. FIG. 9 shows an example of a cost map in which costs are assigned according to the position in the width direction of the travel path. As shown in the figure, the cost is set smaller toward the center of the travel path (the position of the center line) and higher the closer to the boundary line of the travel path. Because passing outside the boundary line is not permitted, a sufficiently large cost Cmax is set at the boundary line position. Furthermore, the route identification unit 21 identifies the optimal travel path using a new cost map in which the numerical values ​​of the cost map are changed to values ​​that take into account a margin width according to the curvature value of the new route point sequence.

[0073] The cost map does not need to be symmetrical in the width direction with the center line position as the axis of symmetry, as shown in Figure 9. Various types of road information are associated with the center point. Road information includes, for example, traffic restrictions (speed limit, one-way street, height limit, width limit, weight limit, etc.), altitude (or elevation difference from a predetermined position), whether the road is paved, and the location of obstacles. Using this information, a cost map in the width direction of the road 30 is generated for each center point position. For example, the cost may be increased in areas with elevation differences, areas with obstacles, unpaved areas, etc. Furthermore, for areas with height or width restrictions, a cost map with different costs in the width direction or height direction may be used.

[0074] The cost map may have cost distributions based on the positions of the centerline and boundary lines, either independently or in association with each other. The cost distribution may be calculated using a cost function. For example, if the road width is sufficiently wide, without considering the centerline, the cost distribution will be large only at both ends of the road width and flat elsewhere. If a driving route is evaluated using such a cost distribution, costs will be low even if the road width is not near the center, and the optimal route will not pass through the center. Therefore, a cost map with a cost distribution in which costs increase as the distance from the centerline to the ends of the road width increases. This makes it easier to select a route near the center of the road as the optimal route, which meets user expectations. Furthermore, at intersections, if the centerline is redrawn using the temporary closing line described below, the optimal route will be selected as the same as the route a human-operated vehicle would take when passing through the intersection.

[0075] Between waypoints in a uniform section, the optimal driving route converges to approximately the same position in the width direction of the roadway for each vehicle. On the other hand, in a non-uniform section, the optimal driving route differs depending on the vehicle's turning radius or wheelbase length. Therefore, it is preferable to search for the optimal driving route for each vehicle. In this case, the driving route of the vehicle with the largest turning radius may be determined, and for other vehicles, the driving route of the vehicle with the largest turning radius may be used as the initial set route to search for the optimal route.

[0076] The above-described method identifies an optimal driving route from an arbitrary start point S to a destination point G. However, the driving route from waypoint A to waypoint B diverges toward the non-traveled road at the intersection. This is because the driving cost is lower in the direction of the wider road, so the optimal route is pulled toward the non-traveled road at the intersection. Specifically, as shown in 1001 of FIG. 10, the center line CL31a of the road from waypoint A on road 30b to waypoint B on road 30a passes through the center of the intersection, which is on the side of the non-traveled road 30c. Generally, the driving cost is lower closer to the center line. Therefore, the optimal driving route tends to be pulled toward the center line. Therefore, it is preferable to generate a center line that excludes the influence of the non-traveled road and evaluate the driving cost.

[0077] The cutoff setting unit 24 restricts entry and exit at an intersection where multiple travel paths 30 intersect. Specifically, for example, as shown in 1002 in FIG. 10 , the cutoff setting unit 24 sets a virtual temporary cutoff line SB31 at the entrance of a travel path 30C that is not included in the route AB identified by the route identification unit 21. The temporary cutoff line SB31 may be referred to as a temporary boundary line. When a center line CL31a is generated with the temporary cutoff line SB31 set, a gently curved center line CL31a that is not pulled toward the travel path 30c is obtained, as shown in 1002 in FIG. 10 . When a travel path is calculated using a cost map based on this center line CL31a, a gently curved travel path is obtained.

[0078] The uniform section extraction unit 22 extracts a uniform section in a state where entry into some of the travel paths 30 at the intersection is restricted by the cutoff setting unit 24. The via point setting unit 23 sets via points at the end regions of the uniform section extracted in a state where entry into some of the travel paths 30 at the intersection is restricted. The route setting unit 20 sets a travel route that passes through the via points set in this way. The route identification unit 21 calculates the travel cost of the set travel route and identifies the optimal travel route.

[0079] By processing in this way, even points that are actually intersections are extracted as uniform sections, and the driving route that passes through them will also be a route with fewer curves, as shown in 1002 in Figure 10. Even if a section is not extracted as a uniform section, the wide space at the intersection can be eliminated, resulting in a smoother route being generated. Furthermore, by closing off driving routes that do not need to be considered for passing through in advance with temporary closing lines, unnecessary routes will not be set, and the cost of route search can be reduced.

[0080] An example of how to use temporary cutoff lines is as follows. Temporary cutoff lines are set in advance at branching points (intersections) on the travel path 30, and each temporary cutoff line can be set to be enabled or disabled. FIG. 11 shows the positions where temporary cutoff lines are generated at intersections, indicated by dotted lines. As shown in FIG. 11, temporary cutoff lines can be set at three locations (see 1101) at a three-way intersection and eight locations (see 1102) at a four-way intersection. When searching for a route, all temporary cutoff lines are disabled. To identify the optimal route, once the section between via points to be used for route generation has been identified, temporary cutoff lines at the entrances of the travel path 30 that are not used for that section are enabled. In this state, the uniform section extraction unit, via point setting unit, route setting unit 20, and route identification unit 21 perform the above-described processing. This allows for the generation of a smoother travel route.

[0081] Temporary deadline lines can be set, for example, using the following method. First, the information processing device 1A is instructed to draw all boundaries without distinguishing between temporary deadline lines and boundary lines. After drawing, the information processing device 1A is instructed to recognize the edges of the travel path. The user may select any points that the information processing device 1A cannot recognize. Some closed boundaries without edges appear, so the boundary lines are recognized as not closed based on their shape, area, etc. These operations make it possible to distinguish between boundary lines and temporary deadline lines. Next, an ID is assigned to the boundary line, and the IDs of the boundary lines connecting to the temporary deadline line are stored as attribute information of the temporary deadline line. For example, if a temporary deadline line is connected to boundary lines IDs 1 and 2, the IDs of the temporary deadline line are set to 1 and 2. When extracting a boundary line for a route, if the boundary line IDs include 1 and 2, the temporary deadline line is enabled. Through the above processing, temporary deadline lines can be set and switched between enabled and disabled.

[0082] More specifically, the deadline setting unit 24 may generate a temporary deadline line as follows. It is assumed that a boundary line that does not include a temporary deadline line has already been generated. Taking advantage of the fact that center point pairs are interchanged at intersections, a line connecting the tangent points on the boundary line of an inscribed circle inscribed in the boundary line of the intersection is generated as a candidate temporary deadline line. Reference numeral 1101 in FIG. 11 illustrates a method for generating a temporary deadline line in the case of a three-way intersection with a single lane (a common lane for both inbound and outbound traffic). In the case of a three-way intersection, the inscribed circle R is as shown in the figure, so the tangent point between the inscribed circle R and boundary line BL1 is R1, the tangent point between the inscribed circle R and boundary line BL2 is R2, and the tangent point between the inscribed circle R and boundary line BL3 is R3. In this case, a line segment SL1 connecting R1 and R3, a line segment SL2 connecting R1 and R2, and a line segment SL3 connecting R2 and R3 are generated as temporary deadline lines.

[0083] At a three-way intersection, the temporary cutoff line can be identified using only the boundary line ID (boundary line identifier). However, at an intersection with four or more intersections, the temporary cutoff line cannot be identified using only the boundary line ID. This is because the travel lanes (sections) connected by the temporary cutoff line cannot be identified using only the boundary line ID.

[0084] Therefore, the deadline setting unit 24 may generate a temporary deadline line as follows. First, the deadline setting unit 24 obtains a combination of two section IDs (section identifiers) from the intersection adjacency information. Next, the deadline setting unit 24 obtains the boundary line IDs held by the two section IDs, and connects the two boundary lines where the non-shared boundary line IDs change, to create a candidate temporary deadline line. The two boundary line IDs and the two section IDs obtained at this time can be stored in the deadline setting unit 24. As a result, it becomes possible to switch the temporary deadline line according to the route, even for intersections with three or more intersections. In this way, the deadline setting unit 24 stores the section ID as an attribute value of the temporary deadline line, making it possible to switch the temporary deadline line.

[0085] In the case of a four-way intersection, eight temporary cutoff lines SL1 to SL8 are generated, as shown by 1102 in Fig. 11. This is because it is necessary to separately generate a temporary cutoff line SL6 for traveling between travel paths 30a and 30b, a temporary cutoff line SL5 for traveling between travel paths 30c and 30d, and to separately generate a temporary cutoff line SL8 for traveling between travel paths 30b and 30c, and a temporary cutoff line SL7 for traveling between travel paths 30a and 30d. In other words, the cutoff setting unit 24 generates temporary cutoff lines by specifying the travel paths connected by the temporary cutoff lines.

[0086] If the information about the two travel lanes connected by a cutoff line is defined as the cutoff line's "section identifier," and the information about the two boundary lines connected by the cutoff line is defined as the cutoff line's "boundary line identifier," then all cutoff lines can be identified using two identifiers: the "section identifier" and the "boundary line identifier." For example, cutoff line SL6 has a "section identifier" of (30a, 30b) and a "boundary line identifier" of (BL1, BL3), so cutoff line SL6 is identified as "cutoff line (30a, 30b) - (BL1, BL3)." Note that when cutoff lines are generated using the method described in 1101 of FIG. 11, the cutoff lines SL6 and SL5, and the cutoff lines SL8 and SL7, are each generated as the same straight line. Therefore, the user may distinguish the two cutoff lines by adding a curve, as shown in 1102 of FIG. 11. When setting a temporary deadline line to be valid or invalid, the deadline setting unit 24 specifies and sets two identifiers: a "section identifier" and a "boundary line identifier." Because the deadline setting unit 24 sets the temporary deadline line as described above, the user does not need to set the temporary deadline line himself.

[0087] FIG. 12 illustrates the concept of treating a four-way intersection with a single lane as an integrated intersection X. An integrated intersection is an intersection where four or more lanes intersect and is treated as an intersection consisting of multiple three-way intersections. A three-way intersection in an integrated intersection indicates a point where the pair of boundary line identifiers (boundary line IDs) held by each of the center points constituting a center point pair changes. This point has the geometric characteristic of a change in the pair of boundary line identifiers held by the center point pair. A center point ID is associated with (holds) the boundary line identifier that served as the basis for generating that center point. The two boundary line identifiers held by the two center point IDs constituting the center point pair are referred to as a "pair of boundary line identifiers." The intersection indicated by 1201 in FIG. 12 is an intersection where lanes (traffic lanes) 21a, 21b, 21c, and 21d defined by four boundary lines 20a, 20b, 20c, and 20d intersect. This intersection can be treated as an integrated intersection X consisting of a three-way intersection Y1 and a three-way intersection Y2. Hereinafter, the three-way intersection will be denoted by a symbol Y and referred to as a "three-way intersection Yx" or simply as an "intersection Yx." Intersections Y1 and Y2 are treated as intersections belonging to one integrated intersection X.

[0088] In the illustrated example, the travel path 21a is a travel path connecting intersections Y1 and Y3. Therefore, the travel path 21a can be identified by a section identifier "Y1-Y3." Intersection Y3 is a three-way intersection adjacent to intersection Y1. Similarly, the travel path 21b is a travel path connecting intersection Y1 and its adjacent intersection Y4, so it can be identified by a section identifier "Y1-Y4." The travel path 21c is a travel path connecting intersection Y2 and its adjacent intersection Y5, so it can be identified by a section identifier "Y2-Y5." The travel path 21d is a travel path connecting intersection Y2 and its adjacent intersection Y6, so it can be identified by a section identifier "Y2-Y6."

[0089] An example of how to determine which intersections should be treated as combined intersections is explained below. First, an intersection with four or more intersections is divided into multiple three-way intersections. Then, based on the relationship between the distance between the multiple three-way intersections and the road width, it can be determined whether or not to treat the intersection as a combined intersection.

[0090] An example of a specific processing method is described below. The information associated with a three-way intersection includes road width information when the center point was generated. Therefore, if at least a portion of the inscribed circles of each three-way intersection overlap, the intersections can be treated as a single integrated intersection. For example, if the inscribed circles A11 of three-way intersection A10, B11 of three-way intersection B10, and C11 of three-way intersection C10 overlap but the inscribed circles A11 and C11 do not overlap, the inscribed circles C11 of three-way intersection A10 and C10 are considered to be contained within a single integrated intersection via the inscribed circle B11. This processing can be performed similarly for N three-way intersections. The user may also appropriately change the ratio of the radius of the inscribed circles used to determine an integrated intersection to determine whether or not to treat it as an integrated intersection.

[0091] The temporary cutoff line holds information using the section ID and boundary line ID between intersections as identifiers. The section IDs between intersections held here are the intersection section IDs of the travel lanes entering the integrated intersection and the intersection section IDs of the travel lanes leaving the integrated intersection, and there are cases where the short section ID between the three-way intersections within the integrated intersection is not held. This is the case when the temporary cutoff line is drawn so that it connects the boundary line ID of the approach road and the boundary line ID of the exit road.

[0092] If the concept of integrated intersections did not exist and only section IDs were held by three-way intersections, it would be necessary to specify all sections from the lane entering the intersection to the short section between the three-way intersections within the intersection, and then specify all sections to the lane exiting. There may be multiple temporary cutoff lines within an intersection, and if the section IDs between the three-way intersections within the intersection are included, unnecessary temporary cutoff lines may be recognized.

[0093] Therefore, when setting temporary cutoff lines, it is sufficient to specify only the lanes entering and leaving the intersection. This is because the temporary cutoff lines are already connected to the boundary lines of the entry and exit lanes. When three-way intersections are close to each other, there is little need to connect temporary cutoff lines to the boundary lines of the short section between the entry and exit lanes. Therefore, temporary cutoff lines connecting to the boundary lines of the entry and exit lanes of a certain route can be set not to be drawn. Furthermore, temporary cutoff lines can be set to only be generated between the boundary lines of the entry and exit lanes. In other words, by using an integrated intersection, the entry and exit lanes of the intersection can be determined, and unnecessary temporary cutoff lines will not be recognized. Furthermore, temporary cutoff lines connecting to the boundary lines of the entry and exit lanes can also be generated by the program based on the information from the integrated intersection.

[0094] In other words, when an intersection is recognized as an integrated intersection, the cutoff lines drawn within that intersection are generated to connect the entry and exit lane boundary lines. On the other hand, if the distance between three-way intersections is far enough, they become independent intersections, so the entry and exit intersection sections and cutoff lines can be set independently. In other words, there is always a correspondence between how cutoff lines are drawn and which three-way intersections are included in the integrated intersection.

[0095] Based on the above assumptions, the temporary cutoff lines to be set at the integrated intersection X can be defined using six intersections Y1 to Y6. For example, when setting a route from intersection Y4 via intersections Y1 and Y2 to intersection Y6, the cutoff line setting unit 24 must set (generate) temporary cutoff lines 10a and 10b. In this case, the identifier "20a" of boundary line 20a and the identifier "20b" of boundary line 20b can be identified from the section information from Y4 to Y1 (travel path 21b). The travel path from Y1 to Y2 is considered to be a travel path defined by boundary lines 20b and 20d. Therefore, the identifier "20b" of boundary line 20b and the identifier "20d" of boundary line 20d can be identified from the section information of that travel path. Next, the identifier "20c" of boundary line 20c and the identifier "20d" of boundary line 20d can be identified from the section information from Y2 to Y6.

[0096] By treating the integrated intersection X as including intersections Y1 and Y2, the identifiers 20a, 20b, 20c, and 20d of the boundary lines 20a, 20b, 20c, and 20d, respectively, can be associated with one integrated intersection X. As a result, the temporary cutoff line 10a at the integrated intersection X can be identified by the boundary line identifier (20b-20c). Similarly, the temporary cutoff line 10b can be identified by the boundary line identifier (20a-20d). By combining this with the section identifiers described above, the temporary cutoff lines 10a and 10b can be specified and set.

[0097] Reference numeral 1202 in FIG. 12 indicates temporary cutoff lines 10c and 10d that are set when turning left or right at integrated intersection X. In this case, too, considering a route that passes through intersections Y1 and Y2 as described above, temporary cutoff line 10c or 10d can be identified using a combination of a section identifier and a boundary line identifier. For example, when setting a route from intersection Y5 to intersection Y6, the route is Y5 → Y2 → Y6, so boundary lines 20b and 20c are first identified from the section information of travel path 21c identified by the section identifier (Y5-Y2). Boundary lines 20c and 20d are identified from the section information of travel path 21d identified by the section identifier (Y2-Y6). In the direction of travel of the travel path, boundary line 20c on the left side is a boundary line used in common by section identifier (Y5-Y2) and section identifier (Y2-Y6), and there are no temporary cut-off lines that have boundary lines 20c and 20c as boundary line identifiers, so it is determined that there is no temporary cut-off line.On the other hand, boundary lines 20b and 20d on the right side are boundary lines used in section identifier (Y5-Y2) and section identifier (Y2-Y6), and there are temporary cut-off lines that have boundary lines 20b and 20d as boundary line identifiers, and it is possible to identify temporary cut-off line 10c by combining the section identifiers and boundary line identifiers of the connected travel paths.

[0098] In summary, at an integrated intersection, the cutoff setting unit 24 identifies two first tangent points where the two boundary lines constituting the approaching travel lane touch an inscribed circle centered on a three-way junction within the integrated intersection. The three-way junction here is as described above. The cutoff setting unit 24 also identifies two second tangent points where the two boundary lines constituting the exiting travel lane touch an inscribed circle centered on a three-way junction within the integrated intersection. Furthermore, the cutoff setting unit 24 connects the first tangent point and the second tangent point on the boundary line on the right side of the travel direction, and connects the first tangent point and the second tangent point on the boundary line on the left side of the travel direction. This allows the cutoff setting unit 24 to restrict entry and exit and generate a temporary cutoff line that creates a single lane for the travel route from the approaching travel lane to the exiting travel lane. Note that if the boundary lines of one of the travel lanes on the approaching and exiting sides are the same, the tangent points on the sides where the boundary line identifier pair is swapped are connected.

[0099] In this embodiment, a point where the pair of boundary line IDs held by the center point changes is called a three-way intersection. This three-way intersection can also be generated in an n-way intersection, with (n-2) three-way intersections being generated within each n-way intersection. Therefore, when defining a general intersection, it can be expressed as being composed of one or more three-way intersections. An integrated intersection is a group of these three-way intersections, and the temporary deadline line can be identified by matching the section ID and boundary line ID of the travel lane entering the integrated intersection with the section ID and boundary line ID of the travel lane leaving the integrated intersection, and the two section IDs held by the temporary deadline line with two sets of boundary line IDs on the left and right sides of the direction of travel.

[0100] If an integrated intersection like the one described above is not used, the temporary cutoff lines that are determined based on which lane to proceed from cannot be identified using the boundary line and lane identifiers. However, by using an integrated intersection, all of the temporary cutoff lines required for the route to be set can be identified using the boundary line and lane identifiers.

[0101] (Information processing method) Next, an information processing method S1 according to this embodiment will be described with reference to the drawings. Fig. 2 is a flowchart showing the flow of the information processing method S1 according to this embodiment. As shown in the figure, the information processing method S1 includes the following steps:

[0102] In step S11, one or more processors (for example, the uniform section extraction unit 22) extract uniform sections of the travel path based on the curvature of the travel path of the vehicle.

[0103] Next, in step S12, one or more processors (for example, the via point setting unit 23) set via points in the end regions of the uniform section.

[0104] Next, in step S13, one or more processors (for example, the route setting unit 20) set a travel route including a travel position in the width direction of the travel path between two adjacent waypoints.

[0105] Next, in step S14, one or more processors (for example, the route identification unit 21) calculate the travel cost of the set travel route and identify the optimal travel route.

[0106] As described above, the information processing device 1A and the information processing method S1 according to this embodiment can set via points even in an irregular road network, and can realize a technology that can identify an optimal driving route that passes through those via points. Furthermore, by configuring the system to be able to set temporary cutoff lines, a smoother driving route can be generated.

[0107] [Embodiment 3] Other embodiments of the present invention will be described below. For ease of explanation, the same reference numerals will be used to designate components having the same functions as those described in the first and second embodiments, and the description thereof will not be repeated.

[0108] In the above-mentioned first and second embodiments, an example was described in which the road has one lane. However, even in the case of a road with multiple lanes, the methods described in the above-mentioned first and second embodiments can be applied by treating each lane as one road. Such embodiments will be described below.

[0109] As shown in Fig. 13, the information processing device 1B according to this embodiment includes an intersection route identification unit 26 in addition to the components of the information processing device 1A according to the second embodiment. Furthermore, when a road has multiple lanes, the boundary line identification unit 11 sets lane dividing lines separating the multiple lanes between the multiple lanes. Furthermore, the boundary line identification unit 11 sets the lane dividing lines as polygons surrounded by vector lines.

[0110] A road with multiple lanes is a road with adjacent lanes that run in opposite directions. In other words, a road with multiple lanes is a compound road with an uphill lane (one or more lanes) and an downhill lane (one or more lanes) that are adjacent to each other with a lane divider in between.

[0111] FIG. 14 is a schematic diagram of a four-way intersection where multiple travel paths intersect. FIG. 14 shows a four-way intersection where multiple travel paths 41, 42, 43, and 44 intersect. The multiple travel path 41 is a multiple travel path in which a travel path 412 leading into the intersection and a travel path 411 leading out of the intersection are adjacent to each other with a lane separator 410 in between. Similarly, the multiple travel path 42 is a multiple travel path in which two travel paths 421 and 422 are adjacent to each other with a lane separator 420 in between. The same applies to the multiple travel paths 43 and 44. Note that these eight travel paths 411, 412, 421, 422, 431, 432, 441, and 442 are travel paths leading up to the intersection.

[0112] Lane separation lines 410, 420, 430, and 440 are each a long, thin rectangular polygon surrounded by vector lines whose directions are indicated by arrows. These vector lines can be treated as boundary lines of the travel lanes. By drawing in this manner, a sequence of center points for each travel lane can be generated using the method described above. For example, travel lane 411 is defined by boundary line 35a and the opposing boundary line 410, so the center point of the inscribed circle tangent to each boundary line is set as the center point of travel lane 411. The same applies to other travel lanes. Furthermore, center point pairs can be generated in the same way, and the points where the combination of the center point pairs changes can be identified as intersections.

[0113] Furthermore, the intersection route identification unit 26 according to this embodiment regards an intersection where multiple travel paths intersect as an integrated intersection X that combines multiple three-way intersections Y, and identifies a travel route from the travel path where the vehicle enters to the travel path where the vehicle exits. The meanings of the three-way intersection Y and the integrated intersection X are as described above.

[0114] As shown in FIG. 14, the integrated intersection X has eight travel paths 411, 412, 421, 422, 431, 432, 441, and 442 connected via six intersections Yx, designated Y1 through Y6. In FIG. 14, double arrows indicate the directions of entry and exit for travel paths 511, 512, 521, 522, 531, 532, 541, and 542 of the travel paths. Note that double arrows are not used for travel paths within an intersection because they may be bidirectional. The intersection route identification unit 26 can identify the travel path of a vehicle by identifying the route that passes through the fewest number of intersections Yx, designated Y1 through Y6, from the travel path that the vehicle entered to the travel path that the vehicle exited. Note that intersections Y7 through Y14 are located on the opposite side of the integrated intersection X for each of the travel paths 411, 412, 421, 422, 431, 432, 441, and 442.

[0115] The integrated intersection X has boundary line identifiers 35a, 35b, 35c, 35d, 410, 420, 430, and 440. The integrated intersection X also has section identifiers for the travel path (section) defined by the two intersections Yx. Therefore, for example, the illustrated temporary cutoff lines 60a, 60b, and 60c are identified by a combination of a boundary identifier and a section identifier. Note that the temporary cutoff lines are not limited to those illustrated, and are set according to the set route.

[0116] For example, consider the case of setting a route from intersections Y8 → Y1 → Y5 → Y6 → Y4 → Y13. In this case, when the entering lane 412 (Y8-Y1) and the exiting lane 441 (Y4-Y13) at the integrated intersection X are specified, two pairs of section identifiers (Y8-Y1) and (Y4-Y13) are identified as temporary cutoff lines 60a and 60c. In other words, it is clear that temporary cutoff lines 60a and 60c must be set. Note that temporary cutoff line 60a is identified using the connected boundary line identifiers (35b-35d) and the section identifiers (Y8-Y1, Y4-Y13). Similarly, temporary cutoff line 60c is identified using the connected boundary line identifiers (410-440) and the section identifiers (Y8-Y1, Y4-Y13).

[0117] As described above, since the integrated intersection X contains information about multiple intersections Yx, by using the integrated intersection X and specifying the intersection identifiers (Yx) or section identifiers of the entrance and exit sides, it is possible to identify the temporary cutoff line without using the adjacent structure of the entrance and exit side intersections Yx.

[0118] Note that when distinguishing between travel directions, for example, 1 or -1 can be assigned to the section identifier. For example, for travel path 412, if travel is permitted in the direction Y8 → Y1, 1 is assigned to the travel path 412, and if travel is prohibited in the direction Y1 → Y8, -1 is assigned to the travel path 412. Also, if travel is permitted in both directions, for example, 0 may be assigned. The reverse direction of Y8 → Y1 is multiplied by -1, so that Y1 → Y8 becomes -1.

[0119] As described above, by introducing the integrated intersection X even in a combined travel road, when the cutoff setting unit 24 sets temporary cutoff lines, as described in the second embodiment, it is possible to identify all temporary cutoff lines by using two identifiers, a "section identifier" and a "boundary line identifier." Even at an intersection with four or more forks or an intersection of combined travel roads, by treating it as an integrated intersection, it is possible to specify the vector line of the lane separation line as one of the "boundary line identifiers."

[0120] As described above, even for roads with multiple lanes, a sequence of center points can be generated using the same method as described in the first and second embodiments. Furthermore, by generating pairs of center points, it is possible to identify points where the combination of the pairs changes as intersections. Furthermore, temporary closing lines that close off portions of roads that intersect at intersections can also be generated in a similar manner. In the above example, a two-lane road with one inbound lane and one outbound lane was used as the composite road. However, the number of lanes is not limited. For example, a four-lane road with two inbound lanes and two outbound lanes may also be used. Furthermore, a left-turn lane or the like can also be treated as a single lane. For these multiple lanes, the method described in the first and second embodiments can be applied by setting the above-mentioned lane separators between the lanes.

[0121] As described in the first to third embodiments, the information processing devices 1, 1A, and 1B repeat a process of generating new information by using information obtained from one series of data processing operations in the next data processing operation. Specifically, the information processing device 1 according to the embodiment identifies boundary lines of a travel lane from map data, derives center points from the identified boundary lines, generates a center point pair from the derived center points, and identifies an intersection from the generated center point pair. Furthermore, the information processing devices 1A and 1B set a temporary cutoff line to connect the two travel lanes between two boundary lines of the boundary lines and travel lanes (section information) that constitute the intersection.

[0122] In other words, the information processing device 1 generates a data structure including boundary line data of the travel path identified from map data by the boundary line identification unit 11, center point data derived by the center point derivation unit 12 using the boundary line data, center point pair data generated by the center point pair generation unit 13 using the center point data, and intersection data generated by the intersection identification unit 14 using the center point pair data. Furthermore, the information processing devices 1A and 1B generate a data structure including: center point data derived by the center point derivation unit 12 using the boundary line data of the travel path and including a boundary line ID as an identifier; center point pair data generated by the center point pair generation unit 13 using the center point data and including the center point ID as an identifier; and intersection data generated by the intersection identification unit 14 using the center point pair data and including the center point pair ID as an identifier. Furthermore, the data structure may further include temporary cutoff line data generated by the cutoff setting unit 24 and including the section ID of the travel path between the intersections and the boundary line ID as identifiers.

[0123] [Software implementation example] The functions of the information processing devices 1, 1A, and 1B (hereinafter referred to as "devices") can be realized by a program that causes a computer to function as the device, and a program that causes a computer to function as each part of the device.

[0124] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0125] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0126] Furthermore, some or all of the functions of the control blocks can be realized by logic circuits. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of the control blocks can also be realized by, for example, a quantum computer.

[0127] Furthermore, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI ​​may run on the control device or on another device (for example, an edge computer or a cloud server).

[0128] 〔summary〕 An information processing device according to aspect 1 of the present invention includes a boundary line identification unit that detects boundary lines defining a roadway from map data and classifies and identifies the boundary lines as those with ends and those without ends. When the roadway has multiple lanes, the boundary line identification unit sets lane dividing lines between the multiple lanes to separate the multiple lanes.

[0129] In the information processing device according to one aspect of the present invention, the boundary line identification unit sets the lane separation line as a polygon surrounded by vector lines.

[0130] An information processing device according to one embodiment of the present invention further includes a uniform section extraction unit that extracts a uniform section of a vehicle's driving path based on the curvature of the driving path; a waypoint setting unit that sets waypoints in end regions of the uniform section; a route setting unit that sets a driving path including a driving position in the width direction of the driving path between two adjacent waypoints; and a route identification unit that calculates the driving cost of the set driving path and identifies an optimal driving path.

[0131] An information processing device according to one embodiment of the present invention further includes an intersection route identification unit that identifies the vehicle's travel route from one of the driving paths to another of the driving paths by regarding the intersection of the driving paths as an integrated intersection that combines multiple three-way intersections, where the pairs of boundary line identifiers held by the center points that make up the center point pair change.

[0132] The information processing device according to one aspect of the present invention further includes a cut-off setting unit that restricts entry and exit at the integrated intersection.

[0133] In an information processing device according to one embodiment of the present invention, the deadline setting unit generates a temporary deadline line that restricts entry and exit at the integrated intersection by identifying two first tangent points where two boundary lines constituting the entry lane touch an inscribed circle centered at the three-way intersection within the integrated intersection, and two second tangent points where two boundary lines constituting the exit lane touch an inscribed circle centered at the three-way intersection within the integrated intersection, and by connecting the first tangent point and the second tangent point on the boundary line on the right side of the direction of travel, and connecting the first tangent point and the second tangent point on the boundary line on the left side of the direction of travel.

[0134] In an information processing device according to one embodiment of the present invention, the deadline setting unit identifies the temporary deadline line using information on the two travel paths connected by the temporary deadline line and information on the two boundary lines connected by the temporary deadline line.

[0135] A data structure according to one embodiment of the present invention includes center point data derived by a center point derivation unit using boundary line data of a roadway, the center point pair data including the center point ID as an identifier, generated by a center point pair generation unit using the center point data, and intersection data generated by an intersection identification unit using the center point pair data, the intersection data including the center point pair ID as an identifier.

[0136] The data structure according to one aspect of the present invention further includes temporary cutoff line data generated by a cutoff setting unit and including, as identifiers, a section ID of the travel path between the intersections and the boundary line ID.

[0137] An information processing method according to one embodiment of the present invention includes the steps of one or more processors detecting boundary lines defining a roadway from map data, classifying and identifying the boundary lines as those with ends and those without ends, and, if the roadway has multiple lanes, setting lane dividing lines separating the multiple lanes between the multiple lanes.

[0138] A program according to one aspect of the present invention is a program for causing a computer to function as the information processing device of aspect 1, and is a program for causing a computer to function as the boundary line identifying unit.

[0139] A non-transitory recording medium according to one aspect of the present invention is a computer-readable non-transitory recording medium having the above program recorded thereon.

[0140] [Additional Notes] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0141] 1, 1A, 1B...Information processing device 11…Border line identification part 12...Center point derivation part 13...Center point pair derivation section 14...Intersection specific section 15...Processor 16...Memory 17...End acquisition part 18...Section information recording section 19...Adjacent information identification unit 20...Path setting section 21...Route identification unit 22...Uniform interval extraction unit 23...Via point setting section 24...Deadline setting section 26...Intersection route identification section 30...Travel path 31a~31e, 32a, 32b...Border line 35a~35e…End line C31a, CP32a...inscribed circle CP31a, CP32a...center point CL31a, CL31b, CL32a...center line

Claims

1. a boundary line specifying unit that detects boundary lines that define a travel path from map data and classifies and specifies the boundary lines as boundary lines with ends and boundary lines without ends; When the road has a plurality of lanes, the boundary line identification unit sets a lane dividing line separating the plurality of lanes between the plurality of lanes, and sets the lane dividing line as a polygon surrounded by vector lines. Information processing device.

2. a uniform section extraction unit that extracts uniform sections of the vehicle's travel path based on the curvature of the travel path; a via point setting unit that sets via points in end regions of the uniform section; a route setting unit that sets a travel route including a travel position in a width direction of the travel path between two adjacent waypoints; a route specification unit that calculates a travel cost of the set travel route and specifies an optimal travel route; The information processing device according to claim 1 , further comprising:

3. 3. The information processing device according to claim 1, further comprising an intersection route identification unit that identifies a vehicle's travel route from one of the driving paths to another of the driving paths by regarding the intersection of the driving paths as an integrated intersection that combines multiple three-way intersections, where the pairs of boundary line identifiers held by the center points that make up the center point pair change.

4. The information processing device according to claim 3 , further comprising a cut-off setting unit that limits entry and exit at the integrated intersection.

5. The information processing device described in claim 4, wherein the deadline setting unit generates a temporary deadline line that restricts entry and exit at the integrated intersection by identifying two first tangent points where two boundary lines constituting the entry side of the travel path touch an inscribed circle centered at the three-way intersection within the integrated intersection, and two second tangent points where two boundary lines constituting the exit side of the travel path touch an inscribed circle centered at the three-way intersection within the integrated intersection, and connecting the first tangent point and the second tangent point on the boundary line on the right side of the travel direction, and connecting the first tangent point and the second tangent point on the boundary line on the left side of the travel direction.

6. The information processing device according to claim 5 , wherein the deadline setting unit specifies the temporary deadline line using information on two travel paths connected by the temporary deadline line and information on the two boundary lines connected by the temporary deadline line.

7. one or more processors detecting boundary lines defining a travel path from map data, and classifying and identifying the boundary lines into boundary lines with ends and boundary lines without ends; When the road has a plurality of lanes, setting lane dividing lines separating the plurality of lanes between the plurality of lanes; Including, In the step of setting the lane dividing lines, the lane dividing lines are set as polygons surrounded by vector lines. Information processing methods.

8. 2. A program for causing a computer to function as the information processing device according to claim 1, the program causing the computer to function as the boundary line specifying unit.

9. A computer-readable non-transitory recording medium on which the program according to claim 8 is recorded.

10. a boundary line identifying unit that detects boundary lines that define a road from map data and classifies and identifies the boundary lines as boundary lines with ends and boundary lines without ends; an intersection route identification unit that identifies a travel route of a vehicle from one of the travel routes to another of the travel routes by regarding the intersection of the travel routes as an integrated intersection that combines multiple three-way intersections, where the intersections are points where pairs of boundary line identifiers held by center points that make up a center point pair change, When the road has a plurality of lanes, the boundary line identification unit sets lane dividing lines separating the plurality of lanes between the plurality of lanes. Information processing device.

Images