Dead angle area extraction method, travel control method, program, dead angle area extraction device, and mobile body

Abstract

To accurately extract a dead angle area located behind an object.SOLUTION: A dead angle area extraction method is a dead angle area extraction method of extracting a dead angle area from a measurement reference point. The dead angle area is an area hidden behind an object as viewed from the measurement reference point in a ground surface around the measurement reference point. The dead angle area extraction method includes: distance information acquisition step S20 of acquiring distance information indicating a distance from the measurement reference point to the object in a three-dimensional space around the measurement reference point; height information acquisition step S30 of acquiring height information indicating a height of the ground surface; and extraction step S40 of extracting the dead angle area based on a relative position between the object detected from the distance information and the measurement reference point, and height information of a measurement target point on the ground surface located behind the object as viewed from the measurement reference point among the ground surface.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a blind spot area extraction method, a driving control method, a program, a blind spot area extraction device, and a moving body. [Background technology]

[0002] Currently, research into autonomous driving of moving bodies such as automobiles is actively progressing (see Patent Document 1, etc.). In autonomous driving, it is necessary to detect pedestrians, other vehicles, etc. on the road surface. The invention described in Patent Document 1 attempts to accurately detect the height of pedestrians, other vehicles, etc. on the road surface based on the height detected by a sensor, the height of the road surface, and the pitch angle of the vehicle itself. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-19445 Summary of the Invention [Problem to be solved by the invention]

[0004] The invention described in Patent Document 1 allows for accurate detection of the height of an object such as another vehicle, but does not describe a method for extracting a blind spot caused by the object. A blind spot is an area hidden behind an object when viewed from the vehicle's perspective. For example, a moving vehicle may be present in the blind spot. A moving vehicle present in the blind spot may suddenly emerge from the blind spot and rapidly approach the vehicle. For this reason, automated driving requires not only object detection but also accurate extraction of the blind spot.

[0005] The present invention has been made in consideration of the above-mentioned problems, and has as its object to provide a blind spot area extraction method and the like that can accurately extract a blind spot area located behind an object. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, a blind spot area extraction method according to one embodiment of the present invention is a blind spot area extraction method that extracts a blind spot area from a measurement reference point, wherein the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as viewed from the measurement reference point, and the blind spot area extraction method includes a distance information acquisition step that acquires distance information indicating the distance from the measurement reference point to the object in three-dimensional space around the measurement reference point, a height information acquisition step that acquires height information indicating the height of the ground, and an extraction step that extracts the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point.

[0007] In addition, in order to achieve the above-mentioned object, a driving control method according to one aspect of the present invention includes the above-mentioned blind spot area extraction method and a driving control step of controlling the driving of a moving body based on the blind spot area extracted by the blind spot area extraction method.

[0008] In order to achieve the above object, a program according to one aspect of the present invention is a program for causing a computer to execute the above-described method for controlling autonomous driving.

[0009] Furthermore, in order to achieve the above-mentioned object, a blind spot area extraction device according to one embodiment of the present invention is a blind spot area extraction device that extracts a blind spot area from a measurement reference point, wherein the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as viewed from the measurement reference point, and the blind spot area extraction device includes: a distance information acquisition unit that acquires distance information indicating the distance from the measurement reference point to the object in three-dimensional space around the measurement reference point; a height information acquisition unit that acquires height information indicating the height of the ground; and an extraction unit that extracts the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point.

[0010] In addition, in order to achieve the above-mentioned object, a moving body according to one aspect of the present invention includes the above-mentioned blind spot area extraction device and a driving control unit that controls the driving of the moving body based on the blind spot area extracted by the blind spot area extraction device.

[0011] These comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a non-transitory computer-readable recording medium such as a CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium. [Effects of the Invention]

[0012] According to the blind spot area extraction method and the like of the present invention, the blind spot area located behind an object can be extracted with high accuracy. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a block diagram showing an example of the configuration of a blind spot area extraction device and a moving body according to an embodiment; [Figure 2] FIG. 10 is a diagram illustrating an example of height information according to the embodiment. [Figure 3] FIG. 10 is a diagram illustrating an example of a grid map used in the obstacle map according to the embodiment. [Figure 4] FIG. 2 is a diagram showing an overview of an obstacle map according to the embodiment. [Figure 5] 1A to 1C are diagrams illustrating an obstacle search method according to an embodiment. [Figure 6] 3 is a flowchart illustrating a blind spot area extraction method and a driving control method according to an embodiment. [Figure 7] 10A and 10B are schematic diagrams illustrating a blind spot area extraction method according to a comparative example. [Figure 8] 10A and 10B are schematic diagrams illustrating problems with a blind spot area extraction method according to a comparative example. [Figure 9] FIG. 1 is a schematic diagram illustrating a blind spot area caused by a slope of the ground. [Figure 10]1A and 1B are schematic diagrams illustrating a blind spot area extraction method according to an embodiment. [Figure 11] 10A and 10B are diagrams illustrating the positional relationship between a reference line and measurement target points on the ground according to the embodiment. [Figure 12] 10 is a schematic diagram showing a first example of a case where a blind spot area is not extracted in the method for extracting a blind spot area according to the embodiment; FIG. [Figure 13] 10 is a schematic diagram showing a second example of a case where a blind spot area is not extracted in the method for extracting a blind spot area according to the embodiment; FIG. [Figure 14] 10A and 10B are schematic diagrams illustrating an example of a case where a blind spot area is extracted in the method for extracting a blind spot area according to the embodiment. [Figure 15] This is an image showing the situation of an intersection as seen from a moving object in the first experiment. [Figure 16] 16 is an image showing the result of applying the Semantic Segmentation technique to the image shown in FIG. 15. [Figure 17] 10 is an obstacle map showing blind spot areas extracted by a blind spot area extraction method according to a comparative example in a first experiment. [Figure 18] 10 is an obstacle map showing blind spot areas extracted by the blind spot area extraction method according to the embodiment in a first experiment. [Figure 19] 10 is an image showing a blind spot area erroneously extracted by a blind spot area extraction method of a comparative example in a first experiment. [Figure 20] 10 is a graph showing the relationship between the precision rate and distance of the extraction results obtained by the blind spot area extraction methods according to the embodiment and the comparative example in the first experiment. [Figure 21] 10 is a graph showing the relationship between the recall rate and distance of the extraction results obtained by the blind spot area extraction methods according to the embodiment and the comparative example in the first experiment. [Figure 22] This is an image showing the situation of an intersection as seen from a moving object in a second experiment. [Figure 23] 23 is an image showing the result of applying the Semantic Segmentation technique to the image shown in FIG. 22. [Figure 24]10 is an obstacle map showing blind spot areas extracted by a blind spot area extraction method according to a comparative example in a second experiment. [Figure 25] 10 is an obstacle map showing blind spot areas extracted by the blind spot area extraction method according to the embodiment in a second experiment. [Figure 26] 10 is a graph showing the relationship between the precision rate and the distance of the extraction results obtained by the blind spot area extraction methods according to the embodiment and the comparative example in the second experiment. [Figure 27] 10 is a graph showing the relationship between the recall rate and distance of the extraction results obtained by the blind spot area extraction methods according to the embodiment and the comparative example in the second experiment. [Figure 28] FIG. 10 is a first diagram illustrating a method for extracting a blind spot area according to a modified example of the embodiment. [Figure 29] FIG. 10 is a second diagram illustrating a method for extracting a blind spot area according to a modified example of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0015] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection configurations, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the present invention. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concepts are described as optional components.

[0016] In addition, each drawing is a schematic diagram and is not necessarily an exact illustration. In addition, the same components are denoted by the same reference numerals in each drawing.

[0017] (Embodiment) A blind spot extraction method according to an embodiment will be described.

[0018] [Basic configuration of blind spot extraction device and moving object] The basic configuration of a blind spot area extraction device and a moving body according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of a blind spot area extraction device 100 and a moving body V10 according to this embodiment.

[0019] The moving body V10 is an autonomously driven moving body that includes a blind spot area extraction device 100 and a driving control unit 50. In this embodiment, the moving body V10 further includes a storage unit 60. The moving body V10 is not particularly limited as long as it is a moving body that can be at least partially controlled by the driving control unit 50. In this embodiment, the moving body V10 is an autonomously driven vehicle.

[0020] The blind spot area extraction device 100 is a device that extracts a blind spot area from a measurement reference point. The blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object when viewed from the measurement reference point. The blind spot area extraction device 100 includes a distance information acquisition unit 20, a height information acquisition unit 30, and an extraction unit 40. In this embodiment, the blind spot area extraction device 100 further includes a measurement unit 10.

[0021] The measurement unit 10 measures the distance from a measurement reference point to an object in a three-dimensional space around the measurement reference point. More specifically, the measurement unit 10 measures the distance from the measurement reference point to a measurement target point, which is a point on the surface of an object located around the measurement reference point. Note that the measurement target point includes not only measurement points whose heights are determined based on the measurement results of the measurement unit 10, but also virtual measurement points whose heights are determined based on height information (described later). Objects that include the measurement target point include, for example, moving objects and stationary objects. For example, moving objects include vehicles and pedestrians, while stationary objects include the ground, including the road surface, stopped vehicles, buildings, signs, etc. In this embodiment, the measurement unit 10 uses the position of a sensor included in the measurement unit 10 as a measurement reference point to measure three-dimensional information, such as the direction (horizontal direction) in which a surrounding object is located, the distance to the surface of the object, and the height, as distance information. For example, a LiDAR (Light Detection and Ranging) or the like can be used as the measurement unit 10. In this embodiment, the measurement unit 10 includes a LiDAR.

[0022] The LiDAR measures the distance between an object and a measurement reference point included in the LiDAR by irradiating the object with laser light and acquiring the laser light reflected from the object. The measurement unit 10 acquires 3D point cloud data indicating the relationship between the horizontal position and height of the object around the measurement reference point from the measured distance, the position of the LiDAR, and the direction of the laser light irradiation.

[0023] The LiDAR emits, for example, 128 laser beams in the vertical direction (i.e., the vertical direction) at different irradiation angles. The vertical viewing angle of the LiDAR is, for example, -25 degrees or more and 15 degrees or less, with the horizontal direction (i.e., the direction perpendicular to the vertical direction) being 0 degrees. The LiDAR can acquire omnidirectional 3D point cloud data by rotating the direction of irradiation of these laser beams by 360 degrees in the horizontal direction. The scan rate of the laser beams is, for example, 5 Hz or more and 20 Hz or less. In this embodiment, the scan rate is set to 10 Hz. That is, in this embodiment, the LiDAR acquires 10 frames of 3D point cloud data per second.

[0024] The distance information acquiring unit 20 acquires distance information indicating the distance from the measurement reference point to an object in three-dimensional space around the measurement reference point. In this embodiment, the distance information acquiring unit 20 acquires the distance information measured by the measuring unit 10.

[0025] The height information acquisition unit 30 acquires height information indicating the height of the ground. The height information includes information corresponding to the vertical height of each point on the ground around the measurement reference point. An example of height information will be described with reference to FIG. 2. FIG. 2 is a diagram illustrating an example of height information according to this embodiment. FIG. 2 is a digital elevation map (DEM) indicating the height of the ground around the measurement reference point P0. The DEM is a map showing height information at each position on the ground, which is partitioned in a matrix. In FIG. 2, the height of each position in FIG. 2 is shown in grayscale. In FIG. 2, the closer the color is to white, the higher the height. Note that positions shown in black in FIG. 2 indicate that there is no height information. In other words, the height of a position shown in black is unknown. The height of such a position whose height is unknown may be estimated from the height of a nearby position. For example, in FIG. 2, the height of the ground is acquired from the height information (DEM) sequentially from the measurement reference point P0 in the direction of the dashed arrow. This process is repeated, and when a position whose height is unknown is reached, the height of that position may be assumed to be equal to the height acquired immediately before. In other words, the height of the position that was acquired immediately before may be used as the height of the position whose height is unknown.

[0026] In this embodiment, the height information acquiring unit 30 acquires height information from the storage unit 60. Note that the method of acquiring height information is not limited to this. For example, the height information acquiring unit 30 may acquire height information generated based on distance information measured by the measurement unit 10.

[0027] The extraction unit 40 extracts a blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point. In this embodiment, the extraction unit 40 detects the position and height of an obstacle protruding upward from the ground. As a result, the extraction unit 40 generates an obstacle map that includes the positions of obstacles located around the measurement reference point. The obstacle map and a method for generating the obstacle map will be described below with reference to FIGS. 3 to 5. FIG. 3 is a diagram showing an example of a grid map used in the obstacle map according to this embodiment. FIG. 4 is a diagram showing an overview of the obstacle map according to this embodiment. FIG. 5 is a diagram explaining an obstacle search method according to this embodiment.

[0028] In this embodiment, the extraction unit 40 generates a 512 x 512 grid map as shown in FIG. 3. The horizontal and vertical dimensions of each grid in the grid map generated in this embodiment are both 0.25 m. The extraction unit 40 determines whether each grid in the grid map has an obstacle, no obstacle, or is unknown, based on the distance information. In this way, an obstacle map as shown in FIG. 4 is generated. The obstacle map includes information regarding the presence or absence of an obstacle in each grid in the grid map. In the obstacle map shown in FIG. 4, grids determined to have an obstacle are shown in black, grids determined not to have an obstacle are shown in white, and grids for which the presence or absence of an obstacle is unknown are shown in gray. Note that grids for which the presence or absence of an obstacle is unknown include blind spots.

[0029] In this embodiment, the extraction unit 40 determines whether or not an obstacle is present based on the angle of the object's surface in a vertical plane including the measurement reference point. The method for determining whether or not an obstacle is present according to this embodiment will be described with reference to FIG. 5.

[0030] In the example shown in FIG. 5, the ground G S A moving object V10 is located at , and an obstacle O is in front of the moving object V10. B The measurement target point P shown in Figure 5 is located TFor example, when determining whether or not there is an obstacle at the measurement target point P T Adjacent to the measurement point P T Measurement target point P is closer to measurement reference point P0 (or located below measurement reference point P0) B and the measurement point P T Adjacent to the measurement point P T Measurement target point P that is farther from measurement reference point P0 (or located above measurement reference point P0) F Using the above, the extraction unit 40 extracts the measurement target point P T 5 is a measurement point whose height is determined based on the measurement result of the measurement unit 10.

[0031] Specifically, the measurement point P B and the measurement point P T The angle of the line segment connecting the two points with the horizontal direction, and the measurement target point P T and the measurement point P F If at least one of the angles of the line segment connecting the measurement target point P T For example, the extraction unit 40 determines that there is an obstacle at the measurement target point P B and the measurement point P T The horizontal distance d B,H , and the vertical distance d B,V and the measurement point P T and the measurement point P F The horizontal distance d F,H , and the vertical distance d F,V These distances and a predetermined angle θ th When the following inequality holds for the measurement point P T It is determined that there is an obstacle.

[0032]

number

[0033] Here, the horizontal direction in Figure 5 and the line segment P T P Fand line segment P B P T The angle θ between F and θ B The following equations (2) and (3) hold true for

[0034]

number

[0035]

number

[0036] Therefore, equation (1) is the angle θ F and θ B At least one of these is a predetermined angle θ th This means that it is more than or equal to this.

[0037] Predetermined angle θ th may be, for example, 45 degrees. In this case, tan θ th is 1.

[0038] As described above, the extraction unit 40 determines whether or not there is an obstacle at each measurement point, and generates an obstacle map based on the determination result. In this embodiment, the blind spot area behind the obstacle is extracted based on the position and height of the obstacle obtained from the obstacle map and the height of the ground behind the obstacle obtained from the height information. The detailed method of extracting the blind spot area used by the extraction unit 40 will be described later.

[0039] The travel control unit 50 controls the travel of the moving body V10. The travel control unit 50 controls the travel of the moving body V10 based on the blind spot area extracted by the blind spot area extraction device 100. For example, the travel control unit 50 controls the operation of the moving body V10, assuming that the moving body will move from the blind spot area extracted by the extraction unit 40 to outside the blind spot area. In other words, the travel control unit 50 assumes that a moving body is in a blind spot area that is not visible from the measurement reference point, and assumes that the moving body will jump out of the blind spot area, and controls the travel of the moving body V10 to avoid contact with the moving body.

[0040] The storage unit 60 stores information for at least one of blind spot area extraction and autonomous driving. In this embodiment, the storage unit 60 stores the above-mentioned height information, map information including information on the road on which the moving body V10 is driving, and the like.

[0041] [Basic configuration of blind spot extraction method and driving control method] The basic configuration of a blind spot area extraction method and a driving control method according to this embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart showing the blind spot area extraction method and driving control method according to this embodiment. The blind spot area extraction method using the blind spot area extraction device 100 according to this embodiment and the driving control method using a moving body V10 will be described below.

[0042] The blind spot area extraction method according to this embodiment is a method for extracting a blind spot area from a measurement reference point, and includes a distance information acquisition step S20, a height information acquisition step S30, and an extraction step S40 shown in Fig. 6. The blind spot area extraction method may also include a measurement step S10. The cruise control method according to this embodiment includes each step included in the blind spot area extraction method and a cruise control step S50.

[0043] 6, in the blind spot area extraction method and driving control method according to this embodiment, first, the measurement unit 10 of the blind spot area extraction device 100 measures the distance from a measurement reference point to an object in a three-dimensional space around the measurement reference point (measurement step S10). In this embodiment, the measurement unit 10 uses a LiDAR to measure, as distance information, three-dimensional information such as the direction (horizontal direction) in which the surrounding object is located, the distance to the surface of the object, and the height, with the position of the sensor included in the measurement unit 10 as the measurement reference point.

[0044] Next, the distance information acquiring unit 20 acquires distance information indicating the distance from the measurement reference point to an object in the three-dimensional space around the measurement reference point (distance information acquiring step S20). In this embodiment, the distance information acquiring unit 20 acquires the distance information measured by the measuring unit 10.

[0045] Next, the height information acquiring unit 30 acquires height information indicating the height of the ground surface (height information acquiring step S30). The height information includes information corresponding to the vertical height of each point on the ground surface around the measurement reference point. In this embodiment, the height information acquiring unit 30 acquires the height information from the storage unit 60.

[0046] Next, the extraction unit 40 extracts a blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point (extraction step S40). In this embodiment, the extraction unit 40 generates an obstacle map and extracts a blind spot area located behind an obstacle based on the position and height of the obstacle obtained from the obstacle map and the height of the ground behind the obstacle obtained from the height information. A detailed method for extracting a blind spot area in extraction step S40 will be described later.

[0047] Next, in the travel control method according to the present embodiment, the travel control unit 50 of the moving body V10 controls the travel of the moving body based on the blind spot area extracted by the blind spot area extraction method (travel control step S50). For example, in the travel control step S50, the travel control unit 50 controls the travel of the moving body V10, assuming that the moving body will move from the blind spot area to the outside of the blind spot area.

[0048] As described above, the blind spot area is extracted by the blind spot area extraction method according to this embodiment, and the driving control method according to this embodiment controls the driving of the moving body V10.

[0049] [Detailed configuration and effects of the blind spot area extraction device and blind spot area extraction method] The detailed configuration and effects of the blind spot area extraction device 100 and the blind spot area extraction method according to this embodiment will be described while comparing with a blind spot area extraction method of a comparative example.

[0050] First, a blind spot area extraction method of the comparative example will be described with reference to FIG. 7. FIG. 7 is a schematic diagram illustrating the blind spot area extraction method of the comparative example. FIG. 7 shows a moving body V90 from which a blind spot area is extracted using the blind spot area extraction method of the comparative example. The upper schematic diagram of FIG. 7 shows the directions of the moving body V90 and each axis in a top view of the ground where the moving body V90 is located. The lower schematic diagram of FIG. 7 shows each axis in a plane (hereinafter also referred to as the rz plane) that passes through measurement reference point P0 and is perpendicular to the ground below measurement reference point P0.

[0051] The y-axis direction shown in the schematic diagram at the top of FIG. 7 is a direction that passes through the measurement reference point P0 and is parallel to the traveling direction of the moving object V90 in a planar view of the ground where the moving object V90 is located. The x-axis direction shown in the schematic diagram at the top of FIG. 7 is a direction that passes through the measurement reference point P0 and is perpendicular to the y-axis direction in a planar view of the ground where the moving object V90 is located. The z-axis direction shown in the schematic diagram at the bottom of FIG. 7 is a direction that passes through the measurement reference point P0 and is perpendicular to the ground where the moving object V90 is located. The r-axis direction shown in the schematic diagram at the bottom of FIG. 7 corresponds to the obstacle search direction described above, and is a direction that is centered on the measurement reference point P0 and is parallel to the xy plane. Note that FIG. 7 shows a diagram in which the x-axis direction and the r-axis direction coincide. In the example shown in FIG. 7, two obstacles O are located on the r-axis. B1 , O B2 exists.

[0052] In the blind spot area extraction method of the comparative example, the blind spot area is extracted using only distance information. Specifically, in the blind spot area extraction method of the comparative example, it is assumed that the ground around the moving object V90 is in the same plane as the ground on which the moving object V90 is located. In other words, it is assumed that the moving object V90 is located on a flat surface and that an obstacle exists on the flat surface. Based on this assumption, the distance between the measurement reference point P0 and the obstacle O is B1 Based on the relative position of the obstacle O B1 The distance d in the r-axis direction of the blind spot area behind U can be calculated.

[0053] That is, as shown in FIG. 7, the height of the measurement reference point P0 from the ground is H0, and the obstacle O B1 When the height of is H1, in the rz plane, it passes through the measurement reference point P0 and B1 The reference line L that is in contact with the upper end of S As shown in Figure 7, the reference line L S If the angle between the axis and the z-axis is φ, then the obstacle O B1 The distance d in the r-axis direction of the blind spot area behind U is expressed by the following equation (4).

[0054]

number

[0055] From the measurement reference point P0, the reference line L S and obstacles B1 Using the distance d1 in the r-axis direction to the point of contact with the angle φ, the following equation (5) holds true:

[0056]

number

[0057] In the blind spot area extraction method of the comparative example, the distance d corresponding to the blind spot area is calculated using the above formulas (4) and (5). U That is, in the blind spot area extraction method of the comparative example, as shown in the upper schematic diagram of FIG. B1 The area behind the obstacle O B1 Distance d from U In the blind spot extraction method of the comparative example, the area within the range of the obstacle O is determined to be a blind spot area. B1 The area behind the obstacle O B1 Distance d from U and is outside the range of obstacle O B2 The area closer to the measurement reference point P0, i.e., the area closer to the obstacle O B1 Blind spots and obstacles behind the B2 The distance d between F In the blind spot area extraction method of the comparative example, the area including the obstacle O is determined to be an area without an obstacle. B2 Regarding the area behind the obstacle O B1 The blind spot area is extracted as well as the area behind the vehicle.

[0058] The problems with the blind spot area extraction method of the comparative example will be described with reference to FIG. 8. FIG. 8 is a schematic diagram for explaining the problems with the blind spot area extraction method of the comparative example. The schematic diagram (a) of FIG. 8 shows the virtual ground G assumed in the blind spot area extraction method of the comparative example. SV The distance d corresponding to the blind spot area in UIn the schematic diagrams (b) and (c) of FIG. 8, the ground G S The shape of the virtual ground G SV The distance d corresponding to the blind spot area when the shape is different from U 8 also shows an area R where distance information can be measured by the measurement unit 10. O (i.e., the area where distance can be measured by LiDAR) is also shown.

[0059] As shown in the schematic diagram (a) of FIG. 8, in the blind spot area extraction method of the comparative example, the ground is a flat virtual ground G including the ground below the measurement reference point P0. SV Therefore, the ground G around the moving body V90 S If the ground is flat, the blind spot area can be extracted with high accuracy. S If the surface is not flat, the blind spot area cannot be extracted with high accuracy.

[0060] For example, as shown in the schematic diagrams (b) and (c), the ground G S If the gradient of the image is different between the area around the moving object V90 and the area away from the moving object V90, the blind spot area cannot be extracted with high accuracy.

[0061] As shown in the schematic diagram (b), in the area away from the moving body V90, the ground G S Even if the height of the vehicle V90 increases as it moves away from the moving object V90, the distance d corresponding to the blind spot area extracted by the blind spot area extraction method of the comparative example is U is the distance d shown in the schematic diagram (a) U However, in reality, as shown in the schematic diagram (b), the distance d U is the distance d shown in the schematic diagram (a) U In this way, in the blind spot region extraction method of the comparative example, S When the shape of the blind spot is as shown in the schematic diagram (b), the distance d U As the actual distance d U Calculate the longer value.

[0062] Also, as shown in the schematic diagram (c), in the area away from the moving body V90, S Even if the height of the vehicle V90 decreases as it moves away from the moving object V90, the distance d corresponding to the blind spot area extracted by the blind spot area extraction method of the comparative example is U is the distance d shown in the schematic diagram (a) U However, in reality, as shown in the schematic diagram (c), the distance d corresponding to the blind spot area is U is the distance d shown in the schematic diagram (a) U In this way, in the blind spot region extraction method of the comparative example, S When the shape of the blind spot is as shown in the schematic diagram (c), the distance d U As the actual distance d U Calculate the shorter value.

[0063] Furthermore, the blind spot extraction method of the comparative example cannot detect blind spots caused by the inclination of the ground. The blind spot caused by the inclination of the ground will be described with reference to Fig. 9. Fig. 9 is a schematic diagram illustrating the blind spot caused by the inclination of the ground.

[0064] As shown in FIG. 9, in an area away from the moving body V90, the ground G S If the height of the ground G in Fig. 9 decreases as it moves away from the moving body V90, S Among them, distance d U As shown in the area indicated by the measurement reference point P0, S There may be blind spots where the object cannot be seen.

[0065] As described above, the blind spot extraction method of the comparative example may not be able to extract the blind spot with high accuracy.

[0066] In the blind spot area extraction device 100 and the blind spot area extraction method according to this embodiment, in order to solve the problems of the blind spot area extraction method of the comparative example as described above, the blind spot area is extracted using not only distance information but also height information. Hereinafter, the detailed configuration of the blind spot area extraction device 100 and the blind spot area extraction method according to this embodiment will be described with reference to Figures 10 and 11.

[0067] FIG. 10 is a schematic diagram illustrating a blind spot area extraction method according to this embodiment. The upper schematic diagram of FIG. 10 shows the directions of the moving body V10 and each axis in a top view of the ground on which the moving body V10 is located. The lower schematic diagram of FIG. 10 shows each axis in a plane that passes through the measurement reference point P0 and is perpendicular to the ground below the measurement reference point P0. The x-axis, y-axis, z-axis, and r-axis shown in FIG. 10 are the same as the x-axis, y-axis, z-axis, and r-axis shown in FIG. 7, respectively. In the example shown in FIG. 10, as in FIG. 7, two obstacles O B1 , O B2 FIG. 11 shows the reference line L S And the ground G S Measurement point P on T1 , P T2 1 is a diagram showing the positional relationship between the

[0068] In the extraction step S40 according to this embodiment, as shown in the schematic diagram at the bottom of FIG. 10, the extraction unit 40 extracts the measurement reference point P0 and the obstacle O1 as seen from the measurement reference point P0. B1 The measurement target point is located behind the ground G below the measurement reference point. S In the extraction step S40, the blind spot area is extracted in a plane (rz plane) perpendicular to the axis of the vehicle. Note that the measurement target points for which it is determined whether or not they are extracted as blind spot areas are virtual measurement points whose heights are determined based on height information.

[0069] In the extraction step S40 according to the present embodiment, when extracting the blind spot area, the ground G S is not assumed to be a flat surface, but the actual ground G based on height information S In the extraction step S40, the extraction unit 40 extracts the blind spot area based on the shape of the obstacle O. B1 The reference line L that is in contact with the upper end of S Specifically, in the extraction step S40, the extraction unit 40 extracts the blind spot area by determining whether the measurement target point is located on the reference line L S If the target point is located below the reference line L, the target point is extracted as a blind spot area. SFor example, in the example shown in FIG. 11, the extraction unit 40 does not extract the measurement target point as a blind spot area when the obstacle O B1 The ground G behind S The two measurement points P above T1 , P T2 The coordinates of each measurement point in the rz plane are acquired based on the height information. Based on these coordinates, the extraction unit 40 extracts the reference line L S The extraction unit 40 calculates the positional relationship between the reference line L S Measurement point P located below T1 is extracted as a blind spot area, and the reference line L S Measurement point P located above T2 is not extracted as a blind spot area. S The vertical relationship between the and the measurement point is determined by the reference line L at the r-axis position of the measurement point. S This refers to the vertical relationship between the z-axis position of the object and the z-axis position of the measurement target point.

[0070] As mentioned above, the ground G S By using the height information, the blind spot area can be extracted with high accuracy.

[0071] In the blind spot area extraction method according to the present embodiment, B1 The area behind the obstacle O B1 Distance d from U and is outside the range of obstacle O B2 The area closer to the measurement reference point P0, i.e., the area closer to the obstacle O B1 Blind spots and obstacles behind the B2 The distance d between F The area spanning is determined to be an area free of obstacles.

[0072] In the extraction step S40 according to the present embodiment, the extraction unit 40 S In this embodiment, the blind spot area due to the inclination of the ground G may be detected. SA method for extracting a blind spot area caused by the inclination of a vehicle will be described with reference to Figs. 12 to 14. Figs. 12 and 13 are schematic diagrams showing a first example and a second example, respectively, of a case where a blind spot area is not extracted by the method for extracting a blind spot area according to this embodiment. Fig. 14 is a schematic diagram showing an example of a case where a blind spot area is extracted by the method for extracting a blind spot area according to this embodiment.

[0073] In the extraction step S40 according to this embodiment, the extraction unit 40 extracts the measurement reference point P0 and the measurement target point P T A line passing through and the measurement point P T At ground G S Based on the angle θ with the normal to the T Specifically, the extraction unit 40 determines whether to extract the measurement target point P T Then, the extraction unit 40 acquires the coordinates of the measurement target point P in three-dimensional space using the height information. T At ground G S For example, the extraction unit 40 extracts the normal of the measurement target point P T the ground G adjacent in the first direction S The coordinates of the first point in the three-dimensional space and the measurement target point P T The ground G adjacent to the second direction (different from the first direction) S The normal can be obtained using the coordinates of the second point in three-dimensional space. T The vector from the first point to the measurement target point P T The cross product of the vector pointing from the first point to the second point may be calculated. This allows the vector in the normal direction to be calculated.

[0074] In this embodiment, the extraction unit 40 extracts the measurement reference point P0 and the measurement target point P T A line passing through and the measurement point P T When the angle θ between the normal to the ground at the measurement point P is less than 90°, T is extracted as a blind spot area, and when the angle θ is 90 degrees or more, the measurement target point P T is not extracted as a blind spot area.

[0075] For example, in the example shown in FIG. 12, the flat ground G S Measurement point P on top T is located, and the measurement point P T is not included in the blind spot area. In this case, the measurement reference point P0 and the measurement target point P T A line passing through and the measurement point P T is greater than 90 degrees, the extraction unit 40 does not extract the measurement target point PT as a blind spot area.

[0076] In the example shown in FIG. 13, the ground G S The height of the measurement point P increases as it moves away from the moving object V10. T The ground G is inclined like this. S If it is located above, the measurement point P T is not included in the blind spot area. In this case, as in the case shown in FIG. 12, the measurement reference point P0 and the measurement target point P T A line passing through and the measurement point P T Since the angle θ between the measurement target point P and the normal to the ground is greater than 90 degrees, the extraction unit 40 T is not extracted as a blind spot area.

[0077] In the example shown in FIG. 14, the ground G S The height of the measurement point P decreases as it moves away from the moving object V10. T The ground G is inclined like this. S If it is located above, the measurement point P T is included in the blind spot area. In such a case, the measurement reference point P0 and the measurement target point P T A line passing through and the measurement point P T Since the angle θ between the measurement target point P and the normal to the ground is smaller than 90 degrees, the extraction unit 40 T is extracted as a blind spot area.

[0078] As described above, in this embodiment, the ground G SIt is also possible to accurately extract blind spots caused by the inclination of the vehicle.

[0079] Furthermore, according to the moving body V10 and driving control method of this embodiment, driving control can be performed based on blind spot areas extracted with high accuracy, making it possible to perform more appropriate driving control.

[0080] [Experimental Results] The following describes experimental results of the blind spot area extraction method (and the blind spot area extraction device 100) according to this embodiment. Below, the results of actually extracting blind spot areas using the blind spot area extraction method according to this embodiment will be described while comparing them with the results extracted using the blind spot area extraction method of the comparative example described above.

[0081] First, the results of the first experiment will be explained using Figs. 15 to 21. Fig. 15 is an image showing the state of an intersection as seen from a moving object V10 in the first experiment. Fig. 16 is an image showing the result of applying Semantic Segmentation to the image shown in Fig. 15. Semantic Segmentation is a method of associating each pixel in an image with a label, category, etc. corresponding to the object, etc., indicated by each pixel. In Fig. 16, an area is divided for each object shown in the image, and each pixel is assigned a color corresponding to the object. Fig. 16 shows an image of a road surface R S 17 and 18 show the blind spot area R extracted by the blind spot area extraction method according to the comparative example and the present embodiment in the first experiment, respectively. U 19 shows the obstacle map in which the blind spot area R is erroneously extracted by the blind spot area extraction method of the comparative example in the first experiment. UF20 is a graph showing the relationship between the precision (Precision) and distance of the extraction results obtained by the blind spot area extraction methods according to the present embodiment and the comparative example in the first experiment. FIG. 21 is a graph showing the relationship between the recall (Recall) and distance of the extraction results obtained by the blind spot area extraction methods according to the present embodiment and the comparative example in the first experiment. The solid line graphs shown in FIGS. 20 and 21 show the precision and recall of the blind spot area extraction method according to the present embodiment, and the dotted line graphs show the precision and recall of the blind spot area extraction method of the comparative example.

[0082] 15 and 16, in the first experiment, a moving object V10 is located in front of an intersection. S In the figure, a vehicle V20 is parked as an example of an obstacle. Also, a road surface R located at the back of the intersection as seen from the moving object V10 is S is inclined, and the road surface R S The height increases.

[0083] In such a situation, as shown in FIGS. 17 and 18, in both the comparative example and the blind spot area extraction method according to the present embodiment, the blind spot area R located behind the vehicle V20 is U However, the blind spot area R extracted by the blind spot area extraction method of the comparative example is U is larger than the blind spot area R extracted by the blind spot area extraction method according to this embodiment. U 17 and 18 were projected onto the coordinate system of the image shown in FIG. 16 in order to evaluate these extraction results. As a result, it was found that the blind spot area extraction method according to the present embodiment can extract blind spot areas with high accuracy, and that the blind spot area extracted by the blind spot area extraction method of the comparative example contains erroneously extracted blind spot area R as shown by the black area in FIG. UF It was found to contain a large amount of

[0084] In order to evaluate the accuracy of the blind spot area extracted by the blind spot area extraction method according to the present embodiment and the comparative example, the precision and recall were calculated. The precision and recall are expressed by the following formulas (6) and (7), respectively.

[0085]

number

[0086]

number

[0087] Here, TP indicates the number of points that are actually included in the blind spot area but are correctly extracted as the blind spot area, FP indicates the number of points that are not actually included in the blind spot area but are incorrectly extracted as the blind spot area, and FN indicates the number of points that are actually included in the blind spot area but are incorrectly not extracted as the blind spot area.

[0088] The precision and recall calculated by the above formulas (6) and (7), respectively, are shown in Figures 20 and 21. The distances shown in Figures 20 and 21 indicate the distance from the moving body V90 shown in Figure 17 and the distance from the moving body V10 shown in Figure 18.

[0089] As shown in Figures 20 and 21, in a range of 40 m or more, including the area behind vehicle V20, the blind spot area extraction method of this embodiment has improved accuracy in extracting blind spots compared to the blind spot area extraction method of the comparative example.

[0090] In the first experiment, the length of the blind spot area extracted by the blind spot area extraction method according to the present embodiment is almost equal to the length of the actual blind spot area, while the length of the blind spot area extracted by the blind spot area extraction method of the comparative example is about 11 m shorter than the length of the actual blind spot area. Such a difference in the length of the blind spot area affects the driving control of the moving object V10. For example, in the first experiment, when the moving object V10 turns right at an intersection, the blind spot area R UOther vehicles hidden in the blind spot area R U The moving body V10 performs driving control so as not to collide with the vehicle even if it jumps out of the blind spot area R. U The size of the blind spot area R U This affects the length of time that the moving object V10 waits until the risk of another vehicle jumping out from the blind spot is reduced. For example, if there is a vehicle moving at a constant speed in the blind spot area, the time that the vehicle will be hidden in the blind spot area can be estimated based on the size of the blind spot area. U If the moving body V90 performs driving control based on the above, the moving body V90 will wait longer than necessary. In contrast, by using the blind spot area extraction method according to the present embodiment, it is possible to prevent the moving body V10 from waiting longer than necessary.

[0091] Next, the results of the second experiment will be explained with reference to Figs. 22 to 27. Fig. 22 is an image showing the state of an intersection as seen from a moving object V10 in the second experiment. Fig. 23 is an image showing the result of applying Semantic Segmentation to the image shown in Fig. 22. As with Fig. 16, the image shown in Fig. 22 also shows the road surface R S 24 and 25 show the blind spot area R extracted by the blind spot area extraction method according to the comparative example and the present embodiment in the second experiment, respectively. U 26 is a graph showing the relationship between the precision and distance of the extraction results obtained by the blind spot extraction methods according to the present embodiment and the comparative example in the second experiment. FIG. 27 is a graph showing the relationship between the recall and distance of the extraction results obtained by the blind spot extraction methods according to the present embodiment and the comparative example in the second experiment.

[0092] 22 and 23, in the second experiment, a moving object V10 is located in front of an intersection. SIn the figure, a vehicle V31 is parked as an example of an obstacle. A vehicle V32 is about to enter the intersection from the rear of the intersection as seen from the moving object V10. A road surface R located at the rear of the intersection as seen from the moving object V10 is also parked. S is inclined, and the road surface R S The height of the

[0093] In such a situation, as shown in FIGS. 24 and 25, in both the comparative example and the blind spot area extraction method according to the present embodiment, the blind spot area R located behind the vehicles V31 and V32 is U However, the blind spot area R extracted by the blind spot area extraction method of the comparative example is U is larger than the blind spot area R extracted by the blind spot area extraction method according to this embodiment. U It is shorter in the vertical direction of each figure.

[0094] In order to evaluate the accuracy of the blind spot areas extracted by the blind spot area extraction methods according to the present embodiment and the comparative example, precision and recall were calculated in the same manner as in the first experimental result. The calculated precision and recall are shown in Figures 26 and 27. The distances shown in Figures 26 and 27 indicate the distance from the moving body V90 shown in Figure 24 and the distance from the moving body V10 shown in Figure 25.

[0095] As shown in Fig. 26, in the range of distances of 60 m or more and 90 m or less, including the areas behind vehicles V31 and V32, the blind spot area extraction method according to this embodiment has a better precision than the blind spot area extraction method of the comparative example. As shown in Fig. 27, in the range of distances of 20 m or more and 60 m or less, including the areas behind vehicles V31 and V32, the blind spot area extraction method according to this embodiment has a better recall than the blind spot area extraction method of the comparative example.

[0096] In this way, the blind spot area extraction method according to this embodiment can improve the accuracy of extracting the blind spot area compared to the blind spot area extraction method of the comparative example.

[0097] In the second experiment, the length of the blind spot area located behind the vehicle V31 extracted by the blind spot area extraction method according to the present embodiment is approximately equal to the length of the actual blind spot area. On the other hand, the length of the blind spot area located behind the vehicle V31 extracted by the blind spot area extraction method of the comparative example is shorter than the length of the actual blind spot area. For example, in the second experiment, the length of the blind spot area R extracted by the blind spot area extraction method of the comparative example is U When the moving body V90 performs driving control based on the blind spot area R, the driving control is performed based on a blind spot area that is shorter than the actual blind spot area, so there may be a case where the time to wait for a vehicle to emerge from the blind spot area is insufficient. U In contrast, by using the blind spot area extraction method according to the present embodiment, the moving object V10 can be detected in the blind spot area R U The vehicle can wait long enough to avoid contact with the vehicle emerging from the area.

[0098] [Variations] A modified example of the blind spot area extraction device 100 and the blind spot area extraction method according to this embodiment will be described with reference to Fig. 28 and Fig. 29. Fig. 28 and Fig. 29 are first and second diagrams, respectively, for explaining the blind spot area extraction method according to the modified example of this embodiment.

[0099] In the extraction step S40 of the blind spot area extraction method according to this embodiment, the extraction unit 40 may extract the measurement target point as a blind spot area if a straight line passing through the measurement reference point and the measurement target point intersects with an object between the measurement reference point and the measurement target point.

[0100] For example, as shown in FIG. 28, the measurement reference point P0 and the measurement target point P Ta A line L passing through a is the measurement reference point P0 and the measurement target point P Ta Between the B1 and intersection point P C etc., the extraction unit 40 extracts the measurement target point P Ta is extracted as a blind spot area. On the other hand, the measurement reference point P0 and the measurement target point P TbA line L passing through b is the measurement reference point P0 and the measurement target point P Tb Since there is no intersection with the object between the measurement target point P Tb is not extracted as a blind spot area.

[0101] Also, as shown in FIG. 29, the measurement reference point P0 and the measurement target point P Tc A line L passing through c is the measurement reference point P0 and the measurement target point P Tc Between the two, the ground G, which is an example of an object, S and intersection point P C etc., the extraction unit 40 extracts the measurement target point P Tc is extracted as a blind spot area.

[0102] The blind spot area extraction device 100 and the blind spot area extraction method according to this modification also make it possible to extract blind spots with high accuracy.

[0103] (Other variations, etc.) Although the blind spot extraction method according to one aspect of the present invention has been described above based on the embodiment, the present invention is not limited to the embodiment. As long as it does not deviate from the spirit of the present invention, various modifications that a person skilled in the art can conceive of to the embodiment may also be included within the scope of the present invention.

[0104] For example, in the above embodiment, an example was given in which the moving body V10 was used as an example of a moving body, but the moving body is not limited to the moving body V10. The moving body may be any object whose running can be controlled.

[0105] Although the blind spot area extraction device 100 according to the present embodiment includes the measurement unit 10, the distance information acquisition unit 20, the height information acquisition unit 30, and the extraction unit 40, the configuration of the blind spot area extraction device 100 is not limited to this. For example, the blind spot area extraction device 100 may include at least one of the driving control unit 50 and the storage unit 60, or may not include the measurement unit 10.

[0106] The following embodiments may also be included within the scope of one or more aspects of the present disclosure.

[0107] (1) Some of the components included in the blind spot area extraction device 100 and the moving body V10 may be a computer system configured with a microprocessor, ROM, RAM, a hard disk unit, etc., or a server connected via a network. A computer program is stored in the RAM or hard disk unit. The microprocessor operates in accordance with the computer program to achieve its function. Here, the computer program is configured by combining multiple instruction codes that indicate instructions to a computer to achieve a predetermined function.

[0108] (2) Some of the components included in the blind spot area extraction device 100 and the moving body V10 may be configured as a single system LSI (Large Scale Integration). The system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip, and specifically, is a computer system configured including a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its functions when the microprocessor operates in accordance with the computer program.

[0109] (3) Some of the components included in the blind spot area extraction device 100 and the mobile body V10 may be configured as an IC card or a standalone module that can be attached to each device. The IC card or the module is a computer system configured with a microprocessor, ROM, RAM, etc. The IC card or the module may include the ultra-multifunctional LSI. The IC card or the module achieves its functions when the microprocessor operates according to a computer program. The IC card or the module may be tamper-resistant.

[0110] (4) Furthermore, some of the components included in the blind spot area extraction device 100 and the moving body V10 may be the computer program or the digital signal recorded on a computer-readable recording medium, such as a flexible disk, a hard disk, a CD-ROM, an MO, a DVD, a DVD-ROM, a DVD-RAM, a BD (Blu-ray (registered trademark) Disc), a semiconductor memory, etc. Alternatively, they may be the digital signal recorded on such a recording medium.

[0111] In addition, some of the components included in the blind spot area extraction device 100 and the mobile body V10 may transmit the computer program or the digital signal via a telecommunications line, a wireless or wired communication line, a network such as the Internet, data broadcasting, etc.

[0112] (5) The present disclosure may be embodied as the blind spot extraction method or driving control method described above. It may also be embodied as a computer program for implementing the blind spot extraction method or driving control method on a computer, or as a digital signal comprising the computer program. Furthermore, the present disclosure may be realized as a non-transitory computer-readable recording medium, such as a CD-ROM, on which the computer program is recorded.

[0113] (6) The present disclosure may also be a computer system having a microprocessor and a memory, the memory storing the computer program, and the microprocessor operating in accordance with the computer program.

[0114] (7) The program or the digital signal may also be implemented by another independent computer system by recording it on the recording medium and transferring it, or by transferring the program or the digital signal via the network, etc.

[0115] (8) The above-described embodiments and modifications may be combined with each other.

[0116] (Addendum) Furthermore, the above description discloses the following techniques.

[0117] (Technology 1) A blind spot area extraction method for extracting a blind spot area from a measurement reference point, wherein the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as viewed from the measurement reference point, the blind spot area extraction method including: a distance information acquisition step for acquiring distance information indicating the distance from the measurement reference point to the object in a three-dimensional space around the measurement reference point; a height information acquisition step for acquiring height information indicating the height of the ground; and an extraction step for extracting the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point.

[0118] (Technology 2) A blind spot area extraction method according to Technology 1, in which in the extraction step, it is determined whether or not to extract the measurement target point as the blind spot area based on the angle between a straight line passing through the measurement reference point and the measurement target point and a normal line to the ground at the measurement target point, and the normal line is acquired based on the height information.

[0119] (Technology 3) A blind spot area extraction method according to Technology 1, wherein in the extraction step, if a line segment passing through the measurement reference point and the measurement target point intersects with the object between the measurement reference point and the measurement target point, the measurement target point is extracted as the blind spot area.

[0120] (Technology 4) The blind spot area extraction method according to any one of Techniques 1 to 3, wherein the extraction step detects the position and height of an obstacle protruding upward from the ground.

[0121] (Technology 5) A blind spot area extraction method according to Technology 4, wherein in the extraction step, if the measurement target point is located below a reference line that passes through the measurement reference point and the measurement target point located behind the obstacle as viewed from the measurement reference point, and is perpendicular to the ground below the measurement reference point, the measurement target point is extracted as the blind spot area, and if the measurement target point is located above the reference line, the measurement target point is not extracted as the blind spot area.

[0122] (Technology 6) A driving control method including the blind spot area extraction method according to any one of technologies 1 to 5, and a driving control step of controlling the driving of a moving body based on the blind spot area extracted by the blind spot area extraction method.

[0123] (Technology 7) A program for causing a computer to execute the blind spot area extraction method according to any one of Techniques 1 to 5 or the driving control method according to Technique 6.

[0124] (Technology 8) A blind spot area extraction device that extracts a blind spot area from a measurement reference point, wherein the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as viewed from the measurement reference point, and the blind spot area extraction device includes: a distance information acquisition unit that acquires distance information indicating the distance from the measurement reference point to the object in three-dimensional space around the measurement reference point; a height information acquisition unit that acquires height information indicating the height of the ground; and an extraction unit that extracts the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point.

[0125] (Technology 9) A moving body comprising the blind spot area extraction device according to Technology 8, and a driving control unit that controls the driving of the moving body based on the blind spot area extracted by the blind spot area extraction device. [Industrial Applicability]

[0126] An autonomous driving control method according to one aspect of the present invention can be applied to, for example, a vehicle for autonomous driving. [Explanation of symbols]

[0127] 10. Measurement section 20 Distance information acquisition section 30 Height information acquisition unit 40 Extraction part 50 Travel control unit 60 Storage section 100 Blind spot area extraction device G S ground L S Reference Line O B , O B1 , O B2 Obstacles P0 Measurement reference point P B , P F , P T , P T1 , P T2 , P Ta , P Tb , P Tc Measurement target point V10, V90 mobile object V20, V31, V32 vehicles

Claims

1. A blind spot area extraction method for extracting a blind spot area from a measurement reference point, comprising: the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as seen from the measurement reference point, The blind spot area extraction method includes: a distance information acquisition step of acquiring distance information indicating a distance from the measurement reference point to the object in a three-dimensional space around the measurement reference point; a height information acquisition step of acquiring height information indicating the height of the ground; and an extraction step of extracting the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object as viewed from the measurement reference point. Blind spot area extraction method.

2. In the extraction step, it is determined whether or not to extract the measurement target point as the blind spot area based on an angle formed between a line passing through the measurement reference point and the measurement target point and a normal line to the ground at the measurement target point; The normal is obtained based on the height information. The blind spot extraction method according to claim 1 .

3. In the extraction step, when a line segment passing through the measurement reference point and the measurement target point intersects with the object between the measurement reference point and the measurement target point, the measurement target point is extracted as the blind spot area. The blind spot extraction method according to claim 1 .

4. In the extraction step, the position and height of the obstacle protruding above the ground are detected. The blind spot extraction method according to claim 1 .

5. In the extraction step, when the measurement target point is located below a reference line that passes through the measurement reference point and the measurement target point located behind the obstacle as viewed from the measurement reference point and is perpendicular to the ground below the measurement reference point, the measurement target point is extracted as the blind spot area, and when the measurement target point is located above the reference line, the measurement target point is not extracted as the blind spot area. The blind spot area extraction method according to claim 4 .

6. A blind spot area extraction method according to any one of claims 1 to 5; and a travel control step of controlling travel of a moving body based on the blind spot area extracted by the blind spot area extraction method. Driving control method.

7. A method for causing a computer to execute the blind spot area extraction method according to any one of claims 1 to 5. program.

8. A blind spot area extraction device that extracts a blind spot area from a measurement reference point, the blind spot area is an area of ​​the ground around the measurement reference point that is hidden behind an object as seen from the measurement reference point, The blind spot area extraction device includes: a distance information acquisition unit that acquires distance information indicating a distance from the measurement reference point to the object in a three-dimensional space around the measurement reference point; a height information acquisition unit that acquires height information indicating the height of the ground; an extraction unit that extracts the blind spot area based on the relative position between the object and the measurement reference point detected from the distance information and the height information of a measurement target point on the ground that is located behind the object when viewed from the measurement reference point. Blind spot area extraction device.

9. A mobile object, The blind spot area extraction device according to claim 8 ; a driving control unit that controls driving of the moving body based on the blind spot area extracted by the blind spot area extraction device. Mobile object.

Images