Map generation system, map generation method, and map generation program

Abstract

To provide a map generation system which ensures accuracy of a digital map.SOLUTION: A processor of a map generation system which generates a digital map in which a road structure is mapped is configured to: acquire point cloud data Dp constructed by three-dimensional scanning from an infrastructure sensor with respect to the road structure including a travel road surface, in a scanning coordinate system Ci; generate conversion data as the point cloud data Dp converted into a road surface coordinate system of which the coordinate axes are assumed in a normal direction of the travel road surface included in an area of interest Au in the road structure and a pair of orthogonal directions orthogonal to the normal direction, in response to a user input of the area of interest Au to the point cloud data Dp in the scanning coordinate system Ci; and output a digital map in which a mask area which masks the outside of the travel road surface in the road structure is defined, in response to a user input of the mask area to the conversion data of the road surface coordinate system.SELECTED DRAWING: Figure 10

Description

[Technical Field]

[0001] The present disclosure relates to a technology for generating a digital map that maps road structures. [Background technology]

[0002] Patent Document 1 discloses a technology for generating a digital map that maps road structures by using three-dimensional point cloud data constructed from outputs from an on-board sensor. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-124781 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology disclosed in Patent Document 1 requires automatic recognition and deletion of ground points that represent the road surface. However, in three-dimensional point cloud data that uses the output from an on-board sensor that moves with the vehicle, it is difficult to accurately distinguish point clouds that represent road structures outside the road surface from ground points through automatic recognition. This could lead to a decrease in the accuracy of the digital map.

[0005] An object of the present disclosure is to provide a map generation system that ensures the accuracy of a digital map. Another object of the present disclosure is to provide a map generation method that ensures the accuracy of a digital map. Yet another object of the present disclosure is to provide a map generation program that ensures the accuracy of a digital map. [Means for solving the problem]

[0006] The technical means of the present disclosure for solving the problems will be described below. Note that the claims and the reference characters in parentheses in this section indicate the correspondence with the specific means described in the embodiments described later in detail, and do not limit the technical scope of the present disclosure.

[0007] A first aspect of the present disclosure is A map generation system having a processor (12) for generating a digital map (Dm) that maps a road structure (Rc), The processor Acquiring point cloud data (Dp) constructed by three-dimensional scanning from infrastructure sensors (Ri) of a road structure including a traveling road surface (Rs) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data of the scanning coordinate system, converted data (Dr) as point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in the normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting a defined digital map of a mask area (Am) that masks the area outside the road surface in the road structure in response to user input of the mask area relative to transformation data of the road surface coordinate system.

[0008] A second aspect of the present disclosure is A map generation method executed by a processor (12) for generating a digital map (Dm) mapping a road structure (Rc), comprising: Acquiring point cloud data (Dp) constructed by three-dimensional scanning from infrastructure sensors (Ri) of a road structure including a traveling road surface (Rs) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data of the scanning coordinate system, converted data (Dr) as point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in the normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting a defined digital map of a mask area (Am) that masks areas outside the traveled road surface in the road structure in response to user input of the mask area relative to transformation data of the road surface coordinate system.

[0009] A third aspect of the present disclosure is A map generation program stored in a storage medium (10) and including instructions to be executed by a processor (12) for generating a digital map (Dm) that maps a road structure (Rc), Acquiring point cloud data (Dp) constructed by three-dimensional scanning from infrastructure sensors (Ri) of a road structure including a traveling road surface (Rs) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data of the scanning coordinate system, converted data (Dr) as point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in the normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting a defined digital map of a mask area (Am) that masks areas outside the road surface in the road structure in response to user input of the mask area relative to transformation data of the road surface coordinate system.

[0010] As described above, in the first to third aspects, point cloud data constructed by three-dimensional scanning of a road structure including a traveling road surface by infrastructure sensors is acquired in a scanning coordinate system. Therefore, according to the first to third aspects, in response to a user input of an area of ​​interest for the point cloud data in the scanning coordinate system, conversion data is generated as point cloud data converted into a road surface coordinate system. In this case, the road surface coordinate system assumes coordinate axes in the normal direction of the traveling road surface included in the area of ​​interest input by the user in the road structure, and in a pair of orthogonal directions perpendicular to the normal direction. Therefore, the conversion data into the road surface coordinate system can accurately reflect coordinate information related to the traveling road surface.

[0011] Furthermore, according to the first to third aspects, a digital map in which a mask area that masks areas outside the traveling road surface in the road structure is defined is output in response to a user input of the mask area relative to the transformation data of the road surface coordinate system. This allows the digital map to reflect not only the coordinate information related to the traveling road surface as described above, but also coordinate information related to the road structure outside the traveling road surface. As a result, it is possible to ensure the accuracy of the digital map. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram showing the overall configuration of a map generation system according to an embodiment; [Figure 2] FIG. 1 is a plan view illustrating a road structure focused on in one embodiment. [Figure 3] FIG. 1 is a perspective view illustrating a road structure that is the focus of attention in one embodiment. [Figure 4] FIG. 2 is a side view illustrating a road structure that is the focus of attention in one embodiment. [Figure 5] FIG. 2 is a schematic diagram illustrating point cloud data acquired according to an embodiment. [Figure 6] 1 is a block diagram showing a functional configuration of a map generation system according to an embodiment. [Figure 7] 1 is a flowchart illustrating a map generation flow according to one embodiment. [Figure 8] 10 is a flowchart illustrating an interaction processing subroutine according to one embodiment. [Figure 9] 10 is a flowchart illustrating an automated processing subroutine according to one embodiment. [Figure 10] FIG. 10 is a schematic diagram illustrating a dialogue processing subroutine according to an embodiment. [Figure 11] FIG. 10 is a schematic diagram illustrating a dialogue processing subroutine according to an embodiment. [Figure 12] FIG. 10 is a schematic diagram illustrating a dialogue processing subroutine according to an embodiment. [Figure 13]FIG. 10 is a schematic diagram illustrating a dialogue processing subroutine according to an embodiment. [Figure 14] FIG. 10 is a schematic diagram illustrating a dialogue processing subroutine according to an embodiment. [Figure 15] FIG. 10 is a schematic diagram illustrating an automatic processing subroutine according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] An embodiment of the present disclosure will now be described with reference to the drawings. A map generation system 1 of an embodiment shown in Fig. 1 is installed in an external center such as a data analysis center in order to generate a digital map Dm.

[0014] The digital map Dm is digital data obtained by mapping a road structure Rc as shown in Figures 2 to 4 using point cloud information. In particular, the digital map Dm of this embodiment represents a road structure Rc in which the travel direction Rd on the travel road surface Rs is restricted to a specific direction, for example, at road entrances and exits. The digital map Dm is used, for example, for processing to detect vehicles traveling in the wrong direction relative to the travel direction Rd by being map-matched with point cloud data Dp constructed as shown in Figure 5 by three-dimensional scanning from infrastructure sensors Ri attached to the road structure Rc.

[0015] The point cloud data Dp obtained by three-dimensional scanning from the infrastructure sensor Ri is also used to generate the digital map Dm by the map generation system 1. Therefore, the infrastructure sensor Ri employs LiDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging), which is capable of three-dimensionally scanning a scanning target area by irradiating the scanning target area with an infrared light beam within a horizontal angle of view θ and a vertical angle of view ψ shown in Figures 3 and 4.

[0016] The infrastructure sensor Ri outputs a signal for each detection pixel that correlates with the distance to targets, including road structures Rc, within the scanning area. Based on the output signals from the infrastructure sensor Ri, point cloud data Dp can be obtained, as shown in Figure 5, in which three-dimensional position coordinates are individually assigned to each scanning point that makes up the scanning point cloud.

[0017] As shown in Figures 3 and 4, the infrastructure sensor Ri is installed in an orientation in which the optical axis Ro, which is the axis of symmetry for the horizontal angle of view θ and the vertical angle of view ψ, is inclined at an angle less than a right angle with respect to the road surface Rs. As a result, a scanning coordinate system Ci, which is a three-dimensional Cartesian coordinate system, is defined for the infrastructure sensor Ri. The origin coordinate of the scanning coordinate system Ci is defined as the coordinate of the intersection with the optical axis Ro on the detection surface of the infrastructure sensor Ri. The Xi axis of the scanning coordinate system Ci is defined as a coordinate axis aligned with the optical axis Ro of the infrastructure sensor Ri. The Yi axis of the scanning coordinate system Ci is defined as a coordinate axis orthogonal to the Xi axis on the reference coordinate plane XYi, on which the horizontal angle of view θ of the infrastructure sensor Ri is assumed. The Zi axis of the scanning coordinate system Ci is defined as a coordinate axis orthogonal to the Xi and Yi axes on the vertical coordinate plane XZi, which includes the Xi axis and on which the vertical angle of view ψ of the infrastructure sensor Ri is assumed.

[0018] The map generation system 1 shown in FIG. 1 includes a presentation unit 2, an input unit 3, and a control unit 4. The presentation unit 2 presents necessary information to a user who is an operator of the map generation system 1. The presentation unit 2 has at least a liquid crystal panel or an organic EL panel, for example, that displays image data including a digital map Dm. The presentation unit 2 may also be a speaker, for example, that outputs audio related to the display of the digital map Dm. The input unit 3 is at least one of a mouse, a trackball, a keyboard, etc., that can receive input from a user. The presentation unit 2 and the input unit 3 may be configured to jointly provide a GUI (Graphical User Interface) to a processing operator under the control of the control unit 4.

[0019] The control unit 4 includes at least one dedicated computer and is connected to the presentation unit 2 and the input unit 3 via at least one of a LAN (Local Area Network) line, a wire harness, an internal bus, and a wireless communication line.

[0020] The control unit 4 has at least one memory 10 and one processor 12. The memory 10 is at least one type of non-transitory tangible storage medium, such as a semiconductor memory, a magnetic medium, or an optical medium, that non-temporarily stores computer-readable programs and data. The processor 12 includes at least one type of core, such as a CPU, a GPU, or a RISC-CPU.

[0021] In the control unit 4, the processor 12 executes a plurality of instructions included in a map generation program stored in the memory 10 to generate the digital map Dm. As a result, the control unit 4 constructs a plurality of functional blocks for generating the digital map Dm. The plurality of functional blocks constructed in the control unit 4 include a data acquisition block 100, an interactive processing block 110, and an automatic processing block 120, as shown in FIG. 6.

[0022] The data acquisition block 100 acquires point cloud data Dp constructed by three-dimensional scanning from infrastructure sensors Ri. The interactive processing block 110 performs interactive processing, which is part of the process of generating a digital map Dm, through interaction with a user who is the operator of the map generation system 1. The automatic processing block 120 automatically performs automatic processing, which is part of the process of generating a digital map Dm and is parallel to the interactive processing of the interactive processing block 110, without interaction with the user. By combining such interactive processing and automatic processing, the digital map Dm is generated as output data from the map generation system 1.

[0023] A map generation method in which the map generation system 1 performs interactive and automatic processing to generate a digital map Dm through the cooperation of these blocks 100, 110, and 120 is executed according to the map generation flow shown in Figures 7 to 9. This map generation flow is executed in response to a start command from a user who is the operator of the map generation system 1. Note that each "S" in this map generation flow represents a step executed by multiple commands included in the map generation program.

[0024] In S10 of the map generation flow, the data acquisition block 100 acquires point cloud data Dp constructed by three-dimensional scanning of a road structure Rc including a traveling road surface Rs from infrastructure sensors Ri in a scanning coordinate system Ci as shown in Fig. 5. Then, in S20 and S30 following S10 of the map generation flow, an interactive processing subroutine and an automatic processing subroutine are executed in parallel, respectively, as shown in Fig. 7. First, the interactive processing subroutine of S20 will be described below with reference to Fig. 8.

[0025] In S210 of the dialogue processing subroutine, the dialogue processing block 110 displays the point cloud data Dp of the scanning coordinate system Ci acquired in the previous S10 on the presentation unit 2 as shown in Fig. 5. At this time, under the definition of the scanning coordinate system Ci described above, the presentation unit 2 displays the point cloud data Dp in two dimensions as two-dimensional image data extending on the reference coordinate plane XYi that is inclined with respect to the traveling road surface Rs.

[0026] In S210, the point cloud data Dp of the scanning coordinate system Ci is displayed in a gradation pattern in which color attributes change as the position coordinates of the scanning point cloud move away from the origin coordinates of the infrastructure sensor Ri. The gradation pattern in the display of the point cloud data Dp may be a hue change state in which the hue of multiple display colors as the color attribute of each scanning point gradually changes according to the distance from the infrastructure sensor Ri. The gradation pattern in the display of the point cloud data Dp may be a color tone change state in which the color attribute, determined by the lightness and saturation of a single display color, gradually changes according to the distance from the infrastructure sensor Ri. Note that FIGS. 5 and 10-14 schematically illustrate the gradation pattern displays described above and below, substituting dot densities that change.

[0027] 8, in S220 following S210 of the dialogue processing subroutine, the dialogue processing block 110 accepts a user input for an area of ​​interest Au for the displayed point cloud data Dp of the scanning coordinate system Ci via the input unit 3. At this time, an input from a user who is an operator of the map generation system 1 is accepted so that the area of ​​interest Au including the traveling road surface Rs in the road structure Rc is surrounded by a frame-shaped outline in the displayed point cloud data Dp as shown in FIG. 10. Therefore, in the previous S210, at least one of a reception display for accepting the user input for the area of ​​interest Au and an instruction display for instructing the user to input for the area of ​​interest Au may be performed by the presentation unit 2.

[0028] In S220, the dialogue processing block 110 determines whether or not a user input for an area of ​​interest Au has been received as a dialogue between the map generation system 1 and the user who is its operator, as shown in Fig. 8. As long as a negative determination is made as a result, S220 is repeated. On the other hand, if a positive determination is made, the dialogue processing subroutine proceeds to S230.

[0029] Thus, in S230, which is executed in response to user input for the area of ​​interest Au, the interactive processing block 110 generates transformed data Dr by converting the point cloud data Dp from the scanning coordinate system Ci shown in Figures 3 to 5 to the road surface coordinate system Cr shown in Figures 3, 4, and 11. Specifically, in S230, a centroid analysis process is performed based on the position coordinates of the scanning points included within the area of ​​interest Au (i.e., within the contour line) that were input by the user for the point cloud data Dp in the scanning coordinate system Ci before coordinate conversion, and the centroid coordinates Pu that constitute the following equation 1 in the area of ​​interest Au are focused on. At the same time, in S230, a principal component analysis process is performed based on the position coordinates of the scanning points included within the area of ​​interest Au (i.e., within the contour line), and the third principal component Wu that constitutes equation 1 in the area of ​​interest Au is focused on among the first to third principal components. Furthermore, in S230, an approximate plane expressed by Equation 1 using the barycentric coordinate Pu and the third principal component Wu is estimated as the traveling road surface Rs that passes through the barycentric coordinate Pu and is perpendicular to the third principal component Wu. Note that in Equation 1, xi, yi, and zi represent position coordinates on the approximate plane that represent the traveling road surface Rs in the scanning coordinate system Ci.

number

[0030] Therefore, in S230, a road surface coordinate system Cr, which is a three-dimensional Cartesian coordinate system, is defined for the estimated traveling road surface Rs as shown in Figures 3, 4, and 11. At this time, the origin coordinates of the road surface coordinate system Cr are set to the coordinates of the intersection of a perpendicular line, which is dropped from the origin coordinates of the scanning coordinate system Ci along the vector direction of the third principal component Wu to a virtual plane XYr including the traveling road surface Rs, with the road surface Rs.

[0031] As a result, in S230, the Zr axis, which serves as the coordinate axis of the road surface coordinate system Cr, is set on a perpendicular line from the origin coordinate of the scanning coordinate system Ci to the virtual plane XYr, along the vector direction of the third principal component Wu, which is the normal direction to the traveling road surface Rs. At the same time, in S230, the Xr axis, which serves as the coordinate axis of the road surface coordinate system Cr, is set perpendicular to the Zr axis on a projection line obtained by orthogonally projecting the Xi axis of the scanning coordinate system Ci onto the virtual plane XYr, along the vector direction of the third principal component Wu, which is the normal direction to the traveling road surface Rs. Furthermore, in S230, the Yr axis, which serves as the coordinate axis of the road surface coordinate system Cr, is set on the virtual plane XYr, perpendicular to the Zr and Xr axes. As a result, the Xr and Yr axes are assumed to be along a pair of orthogonal directions that are perpendicular to the normal direction of the traveling road surface Rs.

[0032] In accordance with the definition of the road surface coordinate system Cr, in S230, a coordinate conversion process is executed to convert the position coordinates of the scanning points included in the point cloud data Dp in the scanning coordinate system Ci into the position coordinates of the scanning points included in the conversion data Dr in the road surface coordinate system Cr. Therefore, in S230, coordinate conversion parameters between the scanning coordinate system Ci and the road surface coordinate system Cr are generated and stored in the memory 10.

[0033] As shown in Fig. 8, in S240 following S230 of the dialogue processing subroutine, the dialogue processing block 110 displays the converted data Dr of the road surface coordinate system Cr after the coordinate conversion on the presentation unit 2 as shown in Fig. 11. At this time, the presentation unit 2 displays the converted data Dr two-dimensionally as two-dimensional image data spreading on the virtual plane XYr including the traveling road surface Rs under the definition of the road surface coordinate system Cr described above.

[0034] In S240, the converted data Dr of the road surface coordinate system Cr is displayed in a gradation pattern in which the color attribute changes as the position coordinates of the scanning point cloud move away from the origin coordinates of the road surface coordinate system Cr. Since the distance between the origin coordinates of the scanning coordinate system Ci, which are the coordinates of the intersection of the detection surface of the infrastructure sensor Ri with the optical axis Ro, and the origin coordinates of the road surface coordinate system Cr is constant, the display of the converted data Dr can also be said to be a gradation pattern in which the color attribute changes as the position coordinates of the scanning point cloud move away from the infrastructure sensor Ri. The gradation pattern in the display of the converted data Dr may be either a hue change state or a color tone change state, similar to the display of the point cloud data Dp in S210.

[0035] As shown in Fig. 8, in S250 following S240 of the dialogue processing subroutine, the dialogue processing block 110 accepts a user input of a road direction Rd for the transformation data Dr of the displayed road surface coordinate system Cr via the input unit 3. At this time, an input from a user who is an operator of the map generation system 1 is accepted so that the road direction Rd assumed on the road surface Rs is indicated by an arrow shown in Fig. 12 in the displayed transformation data Dr. Therefore, in the previous S240, at least one of a reception display for accepting the user input of the road direction Rd and an instruction display for instructing the user to input the road direction Rd may be performed by the presentation unit 2.

[0036] In S250, the dialogue processing block 110 determines whether or not a user input of the road direction Rd has been received as a dialogue between the map generation system 1 and the user who is its operator, as shown in Fig. 8. As long as a negative determination is made as a result, S250 is repeated. On the other hand, if a positive determination is made, the dialogue processing subroutine proceeds to S260.

[0037] In S260, which is executed in response to the user input of the road direction Rd, the interactive processing block 110 further rotates and transforms the transformation data Dr to match the same direction Rd, as shown in FIG. 13. This rotation and transformation process matches (i.e., coincides) the road direction Rd of the road surface Rs included in the area of ​​interest Au in the road structure Rc along the Xr-axis direction, which is one of the directions orthogonal to the normal direction of the road surface Rs in the road surface coordinate system Cr in this embodiment. Therefore, in S260, the coordinate transformation parameters between the scanning coordinate system Ci and the road surface coordinate system Cr stored in S230 are combined with the rotation transformation parameters of the road surface coordinate system Cr, and the combined data is stored in the memory 10. Note that the combined data of the coordinate transformation parameters and the rotation transformation parameters stored in this manner will hereinafter be referred to as combined transformation parameters between the scanning coordinate system Ci and the road surface coordinate system Cr.

[0038] As shown in Fig. 8, in S270 following S260 of the dialogue processing subroutine, the dialogue processing block 110 displays the rotationally transformed conversion data Dr of the road surface coordinate system Cr on the presentation unit 2 as shown in Fig. 13. At this time, on the presentation unit 2, the conversion data Dr is displayed in two dimensions as two-dimensional image data on a virtual plane XYr including the traveling road surface Rs, in which the traveling road direction Rd overlaps with the Xr axis direction, in a gradation pattern similar to S240.

[0039] 8, in S280 following S270 of the dialogue processing subroutine, the dialogue processing block 110 accepts a user input for a mask area Am through the input unit 3 for the transformation data Dr after the rotation transformation of the displayed road surface coordinate system Cr. At this time, an input is accepted from the user, who is the operator of the map generation system 1, so that the mask area Am, which masks the area outside the traveling road surface Rs in the road structure Rc, is surrounded by a frame-shaped outline shown in FIG. 14 in the displayed transformation data Dr. Therefore, in the previous S270, at least one of a reception display for accepting the user input for the mask area Am and an instruction display for instructing the user to input for the mask area Am may be performed by the presentation unit 2.

[0040] In S280, the dialogue processing block 110 determines whether or not a user input for the mask area Am has been accepted as a dialogue between the map generation system 1 and the user who is its operator, as shown in Fig. 8. As long as a negative determination is made, S280 is repeated. On the other hand, if a positive determination is made, the dialogue processing subroutine proceeds to S290.

[0041] In this way, in S290, which is executed in response to the user's input for the mask area Am, the interactive processing block 110 outputs a defined digital map Dm for the area Am. At this time, as the digital map Dm, point cloud data Dp in the scanning coordinate system Ci is output, to which coordinate information for the mask area Am in the road surface coordinate system Cr is added, along with synthesis transformation parameters between the scanning coordinate system Ci and the road surface coordinate system Cr. Here, particularly in this embodiment, it is preferable that position coordinates for each scanning point on and / or within the contour line be added as coordinate information for the mask area Am. As described above, in S290, the output digital map Dm is stored in memory 10, and the interactive processing subroutine is completed.

[0042] Next, the automatic processing subroutine of S30 shown in Fig. 7 will be described with reference to Fig. 9. In S310 of the automatic processing subroutine, the automatic processing block 120 automatically identifies a structure point cloud Dpq shown in Fig. 5 as a scanning point cloud representing a stationary structure Rcq of the road structure Rc shown in Figs. 2 and 3 in the point cloud data Dp of the scanning coordinate system Ci acquired in the previous S10. At this time, the structure point cloud Dpq is identified by frequency analysis of the scanning point cloud with respect to the separation distance Lp from the infrastructure sensor Ri, as shown in Fig. 15. Therefore, particularly in this embodiment, using the point cloud data Dp acquired over multiple frames in S10, a scanning point cloud in which the cumulative frequency of each detection pixel with respect to the separation distance Lp from the infrastructure sensor Ri falls within an intermediate frequency ΔF, which is set to a range between more than 0% and less than 100% of the total frequency, as shown in Fig. 15(B), is identified as the structure point cloud Dpq.

[0043] As shown in Fig. 9, in S320 of the automatic processing subroutine, the automatic processing block 120 stores in the memory 10 coordinate information, which is the position coordinates of each scanning point constituting the identified structure point group Dpq, as information to be added to the digital map Dm output in S290 of the interactive processing routine described above. When this storage in S320 is executed, the automatic processing subroutine is completed. Then, in S290 of the interactive processing routine, the stored coordinate information of the structure point group Dpq is added to the digital map Dm together with the synthesis transformation parameters between the coordinate systems Ci and Cr and the coordinate information of the mask area Am. Therefore, upon completion of S290 by performing the above-mentioned addition processing, the map generation flow ends.

[0044] (Action and effect) The effects of the present embodiment described above will be explained below.

[0045] In this embodiment, point cloud data Dp constructed by three-dimensional scanning of a road structure Rc including a traveling road surface Rs by infrastructure sensors Ri is acquired in a scanning coordinate system Ci. Therefore, according to this embodiment, in response to a user input of an area of ​​interest Au for the point cloud data Dp in the scanning coordinate system Ci, conversion data Dr is generated as point cloud data Dp converted into a road surface coordinate system Cr. In this case, the road surface coordinate system Cr assumes coordinate axes Xr, Yr, and Zr in the normal direction of the traveling road surface Rs included in the user-input area of ​​interest Au in the road structure Rc and in a pair of orthogonal directions perpendicular to the normal direction. Therefore, the conversion data Dr for the road surface coordinate system Cr can accurately reflect coordinate information regarding the traveling road surface Rs.

[0046] Moreover, according to this embodiment, a digital map Dm, which defines a mask area Am that masks the area outside the traveling road surface Rs in the road structure Rc, is output in response to a user input of the mask area Am for the conversion data Dr of the road surface coordinate system Cr. As a result, the digital map Dm can reflect not only the coordinate information related to the traveling road surface Rs as described above, but also coordinate information related to the road structure Rc outside the traveling road surface Rs. As a result of the above, it is possible to ensure the accuracy of the digital map Dm.

[0047] According to this embodiment, the user input for the area of ​​interest Au is accurate for the point cloud data Dp of the displayed scanning coordinate system Ci, and the accuracy of the coordinate information reflected for the traveling road surface Rs can be ensured in the conversion data Dr generated in response thereto. Moreover, the user input for the mask area Am is also accurate for the conversion data Dr of the displayed road surface coordinate system Cr, and the accuracy of the coordinate information reflected for the traveling road surface Rs as well as the road structure Rc outside the traveling road surface Rs can be ensured in the digital map Dm output in response thereto. Therefore, it is possible to generate the digital map Dm with high accuracy.

[0048] According to this embodiment, point cloud data Dp of a scanning coordinate system Ci constructed by three-dimensional scanning from infrastructure sensors Ri, which is assumed to have a horizontal angle of view θ on a reference coordinate plane XYi that is inclined relative to the road surface Rs, is displayed in a gradation pattern. In this gradation pattern display, color attributes change as the position coordinates of the scanning points in the point cloud data Dp move away from the infrastructure sensor Ri, which can promote accurate user input of an area of ​​interest Au and contribute to highly accurate generation of a digital map Dm.

[0049] According to this embodiment, the converted data Dr of the road surface coordinate system Cr is displayed in a gradation pattern. Even with this gradation pattern display, the color attributes change as the position coordinates of the scanning point cloud in the converted data Dr move away from the infrastructure sensor Ri, which can promote accurate user input of the mask area Am, thereby contributing to highly accurate generation of the digital map Dm.

[0050] According to this embodiment, the transformation data Dr is rotationally transformed in response to a user input of the road direction Rd so that the road direction Rd of the road surface Rs included in the attention area Au in the road structure Rc matches along one orthogonal direction to the normal direction of the road surface Rs in the road surface coordinate system Cr. This can promote accurate user input of the mask area Am for the transformation data Dr in the road surface coordinate system Cr that is to be displayed in a rotationally transformed state, which can contribute to highly accurate generation of the digital map Dm.

[0051] According to this embodiment, the point cloud data Dp of the scanning coordinate system Ci, to which coordinate information of the mask area Am in the road surface coordinate system Cr is added along with composite transformation parameters between the scanning coordinate system Ci and the road surface coordinate system Cr, is output as a digital map Dm. As a result, in a scenario in which the generated digital map Dm is utilized, for example, in map matching with the point cloud data Dp from the infrastructure sensor Ri described above, it becomes possible to accurately detect moving objects limited to the traveling road surface Rs while omitting additional coordinate transformation processing.

[0052] In this embodiment, a structure point cloud Dpq representing a stationary structure Rcq of the road structure Rc is identified by frequency analysis of the scanning point cloud related to the separation distance Lp from the infrastructure sensor Ri in the point cloud data Dp of the scanning coordinate system Ci. Thus, the point cloud data Dp of the scanning coordinate system Ci, to which coordinate information of the structure point cloud Dpq has been added, is output as the digital map Dm, enabling highly accurate generation of a digital map Dm that also accurately reflects the structure point cloud Dpq. Particularly, according to this embodiment, a scanning point cloud whose cumulative frequency related to the separation distance Lp in the point cloud data Dp of the scanning coordinate system Ci is within the intermediate frequency ΔF can be accurately identified as the structure point cloud Dpq with high certainty, thereby contributing to highly accurate generation of the digital map Dm.

[0053] (Other embodiments) Although one embodiment has been described above, the present disclosure should not be construed as being limited to the embodiment described above, and can be applied to various embodiments within the scope that does not deviate from the gist of the present disclosure.

[0054] As a variant, the dedicated computer constituting the map generation system 1 may have at least one of a digital circuit and an analog circuit as a processor. Here, the digital circuit is at least one of the following: an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a system on a chip (SOC), a programmable gate array (PGA), and a complex programmable logic device (CPLD). Such a digital circuit may also have a memory that stores a program.

[0055] As a modified example, in at least one of steps S210, S240, and S270, instead of a gradation display in which the color attribute changes, a monochromatic display in which the color attribute of each scanning point is set constant may be performed. As a modified example, by omitting steps S250, S260, and S270, the user input of the mask area Am in S280 may be accepted with respect to the conversion data Dr of the road surface coordinate system Cr that has been coordinate-converted in S230 and displayed in S240.

[0056] As a modified example, S30 (S310, S320) may be omitted, and the coordinate information of the structure point group Dpq may be omitted from the additional information to the digital map Dm in S290. As a modified example, in S290, the synthesis transformation parameters between the coordinate systems Ci and Cr may be omitted from the additional information to the digital map Dm, and the transformation data Dr of the road surface coordinate system Cr to which the coordinate information of the mask area Am has been added may be output as the digital map Dm.

[0057] In addition to the forms described so far, the above-described embodiments and variations may be implemented in the form of a processing device (e.g., a processing ECU, etc.) or a semiconductor device (e.g., a semiconductor chip, etc.) as a map generation system having at least one processor 12 and one memory 10.

[0058] (Additional remarks) This specification discloses the following technical ideas and their combinations. Note that the reference symbols in parentheses in this appendix indicate the correspondence with the specific means described in the above detailed embodiments, and do not limit the technical scope of the present disclosure.

[0059] (Technical thought 1) A map generation system having a processor (12) for generating a digital map (Dm) that maps a road structure (Rc), The processor: Acquiring point cloud data (Dp) constructed by three-dimensional scanning from infrastructure sensors (Ri) of the road structure including the travel road surface (Rs) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data in the scanning coordinate system, converted data (Dr) as the point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in a normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting the digital map, in which a mask area (Am) that masks areas outside the travel surface in the road structure is defined, in response to user input of the mask area relative to the transformation data of the road surface coordinate system.

[0060] (Technical thought 2) generating the transformation data generating the transformation data in response to user input of the area of ​​interest relative to the point cloud data in the scanned coordinate system that is displayed; outputting the digital map The map generating system according to technical idea 1 includes outputting the digital map in response to user input of the mask area relative to the transformation data of the road surface coordinate system to be displayed.

[0061] (Technical Thought 3) generating the transformation data A map generation system according to Technical Idea 2, which includes displaying the point cloud data of the scanning coordinate system constructed by three-dimensional scanning from the infrastructure sensor, which is assumed to have a horizontal angle of view (θ) on a reference coordinate plane (XYi) that is inclined from the road surface, in a gradation form in which color attributes change as the position coordinates of the scanning point cloud move away from the infrastructure sensor.

[0062] (Technical Thought 4) outputting the digital map A map generation system according to Technical Idea 2 or 3, which includes displaying the transformation data of the road surface coordinate system in a gradation form in which color attributes change as the position coordinates of the scanning point cloud move away from the infrastructure sensor.

[0063] (Technical Thought 5) generating the transformation data A map generation system described in any one of technical ideas 2 to 4, which includes rotating and transforming the transformation data in response to a user input of the road direction so that the road direction (Rd) of the road surface included in the area of ​​interest in the road structure matches along one of the orthogonal directions in the road surface coordinate system.

[0064] (Technical Thought 6) outputting the digital map A map generation system described in any one of technical ideas 1 to 5, which includes outputting the point cloud data of the scanning coordinate system as the digital map, to which coordinate information of the mask area in the road surface coordinate system has been added along with transformation parameters between the scanning coordinate system and the road surface coordinate system.

[0065] (Technical Thought 7) The processor: The method is further configured to identify a structure point cloud (Dpq) representing a stationary structure (Rcq) among the road structures by frequency analysis of the scanning point cloud with respect to a distance (Lp) from the infrastructure sensor in the point cloud data of the scanning coordinate system; outputting the digital map A map generation system according to Technical Idea 6, which includes outputting the point cloud data of the scanning coordinate system to which coordinate information of the structure point cloud has been added as the digital map.

[0066] (Technical Thought 8) Identifying the structure point cloud includes: A map generation system according to Technical Idea 7, which includes identifying a scanning point cloud whose cumulative frequency with respect to the separation distance in the point cloud data of the scanning coordinate system is within a medium frequency (ΔF) as the structure point cloud.

[0067] The above-mentioned technical concepts 1 to 8 may be understood as the respective technical concepts of a method and a program. [Explanation of symbols]

[0068] 1: Map generation system, 10: Memory, 12: Processor, Am: Mask area, Au: Area of ​​interest, Ci: Scanning coordinate system, Cr: Road surface coordinate system, Dm: Digital map, Dp: Point cloud data, Dpq: Structure point cloud, Dr: Conversion data, Lp: Distance, Rc: Road structure, Rcq: Stationary structure, Rd: Road direction, Ri: Infrastructure sensor, Rs: Road surface, XYi: Reference coordinate plane, Xr, Yr, Zr: Coordinate axes, ΔF: Setting frequency, θ: Horizontal angle of view

Claims

1. A map generation system having a processor (12) for generating a digital map (Dm) that maps a road structure (Rc), comprising: The processor: Acquiring point cloud data (Dp) constructed by three-dimensional scanning of the road structure including the travel road surface (Rs) from an infrastructure sensor (Ri) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data in the scanning coordinate system, converted data (Dr) as the point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in a normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting the digital map, in which a mask area (Am) that masks the area outside the traveling road surface in the road structure is defined, in response to user input of the mask area relative to the transformation data of the road surface coordinate system.

2. generating the transformation data generating the transformation data in response to user input of the area of ​​interest relative to the point cloud data in the scanned coordinate system that is displayed; outputting the digital map 2. The map generating system of claim 1, further comprising: outputting the digital map in response to user input of the mask area relative to the transformation data of the road surface coordinate system to be displayed.

3. generating the transformation data 3. The map generation system according to claim 2, further comprising: displaying the point cloud data of the scanning coordinate system constructed by three-dimensional scanning from the infrastructure sensor, which is assumed to have a horizontal angle of view (θ) on a reference coordinate plane (XYi) that is inclined with respect to the road surface, in a gradation form in which color attributes change as position coordinates of the scanning point cloud move away from the infrastructure sensor.

4. outputting the digital map The map generating system according to claim 2 , further comprising displaying the transformation data of the road surface coordinate system in a gradation form in which color attributes change as the position coordinates of the scanning point cloud move away from the infrastructure sensor.

5. generating the transformation data 3. The map generating system according to claim 2, further comprising: rotating and transforming the transformation data in response to a user input of a road direction (Rd) of the road surface included in the area of ​​interest in the road structure so that the road direction Rd is matched along one of the orthogonal directions in the road surface coordinate system.

6. outputting the digital map 3. The map generation system according to claim 1, further comprising: outputting the point cloud data of the scanning coordinate system as the digital map, to which coordinate information of the mask area in the road surface coordinate system is added together with transformation parameters between the scanning coordinate system and the road surface coordinate system.

7. The processor: The method is further configured to identify a structure point cloud (Dpq) representing a stationary structure (Rcq) among the road structures by frequency analysis of the scanning point cloud with respect to a separation distance (Lp) from the infrastructure sensor in the point cloud data of the scanning coordinate system; outputting the digital map The map generating system according to claim 6 , further comprising: outputting the point cloud data in the scanning coordinate system to which coordinate information of the structure point cloud has been added as the digital map.

8. Identifying the structure point cloud includes: The map generation system according to claim 7 , further comprising identifying, as the structure point cloud, a scanning point cloud whose cumulative frequency with respect to the separation distance in the point cloud data of the scanning coordinate system is within a medium frequency (ΔF).

9. A map generation method executed by a processor (12) for generating a digital map (Dm) mapping a road structure (Rc), comprising: Acquiring point cloud data (Dp) constructed by three-dimensional scanning of the road structure including the travel road surface (Rs) from an infrastructure sensor (Ri) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data in the scanning coordinate system, converted data (Dr) as the point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in a normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting the digital map, in which a mask area (Am) that masks areas outside the traveling road surface in the road structure is defined, in response to user input of the mask area relative to the transformation data of the road surface coordinate system.

10. A map generation program stored in a storage medium (10) and including instructions to be executed by a processor (12) for generating a digital map (Dm) that maps a road structure (Rc), the map generation program comprising: Acquiring point cloud data (Dp) constructed by three-dimensional scanning of the road structure including the travel road surface (Rs) from an infrastructure sensor (Ri) in a scanning coordinate system (Ci); generating, in response to a user input of the area of ​​interest (Au) for the point cloud data in the scanning coordinate system, converted data (Dr) as the point cloud data converted into a road surface coordinate system (Cr) in which coordinate axes (Xr, Yr, Zr) are assumed in a normal direction of the traveling road surface included in the area of ​​interest (Au) in the road structure and in a pair of orthogonal directions perpendicular to the normal direction; and outputting the digital map, in which a mask area (Am) that masks areas outside the traveling road surface in the road structure is defined, in response to user input of the mask area relative to the transformation data of the road surface coordinate system.

Images