Position estimation device and position estimation method

Abstract

A position estimation device (20) comprises: a sensor (30) configured so as to detect the distance to points where electromagnetic waves hit; and a control device (21). Among the points, a plurality of points aligned in the traveling direction of a vehicle are detection points. The control device (21) is configured so as to: set, as a road surface determination target point, one of a plurality of points excluding the point closest to the sensor (20); determine whether detection points at positions closer to the sensor (20) than the road surface determination target point are points representing a road surface; derive a straight line passing through two detection points that are located at positions closer to the sensor (20) than the road surface determination target point and have been determined to be points representing the road surface; determine that the road surface determination target point is a detection point representing the road surface when the distance between the straight line and the road surface determination target point is equal to or less than a threshold; and, in a direction away from the sensor (20), sequentially change the road surface determination target point from a point closer to the sensor (20).

Description

Position Estimation Device and Position Estimation Method

[0001] The present disclosure relates to a position estimation device and a position estimation method.

[0002] A vehicle includes a position estimation device for estimating an obstacle that obstructs the progress of the vehicle. The position estimation device derives the three-dimensional coordinates of a point representing a part of an object from the detection result of a sensor. In this type of position estimation device, the road surface may be detected as an object, and it is necessary to distinguish whether the detected object is the road surface or an obstacle. For example, in Patent Document 1, the gradient of the road surface is estimated. By estimating the gradient of the road surface, the road surface and the obstacle can be separated.

[0003] Japanese Patent Application Laid-Open No. 2015-75382

[0004] When separating the road surface and the obstacle using the gradient of the road surface, the gradient of the road surface needs to be uniform. If it is regarded as a road surface with a uniform gradient, the estimated height of the road surface at a position farther from the vehicle will deviate from the actual height of the road surface.

[0005] According to one aspect of the present disclosure, a position estimation device provided in a vehicle is provided. The position estimation device includes a sensor configured to irradiate electromagnetic waves at a predetermined angle with respect to the vertical direction and detect the distance to the point where the electromagnetic waves hit, and a control device. A plurality of the points arranged in the traveling direction of the vehicle among the points are detection points. The control device sets one of the points excluding the point closest to the sensor among the plurality of points arranged in the traveling direction as a road surface determination target point, determines whether the detection point located closer to the sensor than the road surface determination target point represents a point on the road surface, derives a straight line passing through two of the detection points that are determined to be the points representing the road surface and are located closer to the sensor than the road surface determination target point, and when the distance between the straight line and the road surface determination target point is less than or equal to a threshold value, determines that the road surface determination target point represents the point on the road surface, and is configured to sequentially change the road surface determination target point in the direction away from the sensor from the point closer to the sensor.

[0006] Another aspect of this disclosure provides a position estimation method applicable to a vehicle. The position estimation method includes irradiating a sensor with electromagnetic waves at predetermined angles with respect to the vertical and detecting the distance to the point struck by the electromagnetic waves. A plurality of the points aligned in the direction of travel of the vehicle are detection points. The position estimation method further includes setting one of the plurality of points aligned in the direction of travel, excluding the point closest to the sensor, as a road surface determination target point; determining whether the detection point located closer to the sensor than the road surface determination target point represents a point on the road surface; deriving a straight line passing through two detection points that are located closer to the sensor than the road surface determination target point and have been determined to represent the road surface; determining that the road surface determination target point represents the road surface if the distance between the straight line and the road surface determination target point is less than or equal to a threshold; and sequentially changing the road surface determination target points from the point closest to the sensor to the point further away from the sensor.

[0007] Figure 1 is a side view showing a vehicle equipped with a position estimation device. Figure 2 is a flowchart showing the processing performed by the control device of the position estimation device shown in Figure 1. Figure 3 shows an example of the distance between a straight line passing through two detection points and a road surface determination target point. Figure 4 shows an example of road surface reflection points and obstacle points.

[0008] An embodiment of a position estimation device and a position estimation method will be described. As shown in Figure 1, the vehicle 10 comprises a body 11 and wheels 12. The vehicle 10 may be an industrial vehicle or a passenger car. An industrial vehicle is a forklift or a towing tractor. The vehicle 10 may be driven automatically or manually.

[0009] <Position Estimation Device> The vehicle 10 is equipped with a position estimation device 20. The position estimation device 20 estimates the positions of obstacles present around the vehicle 10. The position estimation device 20 comprises a control device 21 and a sensor 30.

[0010] Sensor 30 detects the distance to the point where the electromagnetic wave strikes by irradiating it with electromagnetic waves. Sensor 30 in this embodiment is a laser rangefinder that irradiates laser light as electromagnetic waves. Laser rangefinders are sometimes called LIDAR (Laser Imaging Detection and Ranging). Laser rangefinders may be mechanical scanning type or solid-state type. Sensor 30 may also be a radar that irradiates radio waves as electromagnetic waves.

[0011] Sensor 30 detects the position of an object in a three-dimensional coordinate system. Sensor 30 detects the distance to a point by irradiating the surrounding area with laser light and receiving the reflected light from the point where the laser light strikes. The point where the laser light strikes represents a part of the object's surface. The position of the point can be expressed in polar coordinates. The coordinates of the point in polar coordinates are converted to coordinates in Cartesian coordinates. The conversion from polar coordinates to Cartesian coordinates may be performed by sensor 30 or by a device other than sensor 30. In this embodiment, it is assumed that the conversion from polar coordinates to Cartesian coordinates is performed by sensor 30.

[0012] Sensor 30 derives the coordinates of a point in the sensor coordinate system. The sensor coordinate system is a three-axis Cartesian coordinate system with sensor 30 as the origin. For example, when the vehicle 10 is positioned on a horizontal road surface, the sensor coordinate system has an X-axis extending horizontally in the left-right direction of the vehicle 10, a Y-axis extending horizontally in the front-rear direction of the vehicle 10, and a Z-axis perpendicular to the X and Y axes. Sensor 30 outputs the coordinates of multiple points obtained by irradiating with laser light as point cloud data. In the following description, coordinates refer to the coordinates of the sensor coordinate system.

[0013] The sensor 30 is positioned to irradiate laser light in the direction of travel of the vehicle 10. The direction of travel of the vehicle 10 is either forward or backward. In this embodiment, the sensor 30 is positioned to irradiate laser light forward. For example, the sensor 30 is positioned at the front end of the vehicle body 11.

[0014] Sensor 30 emits laser light at predetermined angles relative to the vertical. Sensor 30 also emits laser light at predetermined angles relative to the horizontal. In this way, sensor 30 obtains the coordinates of a point in a three-dimensional coordinate system by changing the irradiation angle of the laser light in both the vertical and horizontal directions.

[0015] The control device 21 comprises a processor 22 and a storage unit 23. The processor 22 is, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a DSP (Digital Signal Processor). The storage unit 23 includes RAM (Random Access Memory) and ROM (Read Only Memory). The storage unit 23 stores program code or instructions configured to cause the processor 22 to execute processing. The storage unit 23, i.e., the computer-readable medium, includes any available medium that can be accessed by a general-purpose or dedicated computer. The control device 21 may be composed of hardware circuits such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The control device 21, which is a processing circuit, may include one or more processors that operate according to a computer program, one or more hardware circuits such as an ASIC or FPGA, or a combination thereof.

[0016] The control device 21 estimates the position of obstacles in the direction of travel of the vehicle 10. The point cloud data output by the sensor 30 includes points representing the road surface and points representing obstacles. Points representing the road surface are obtained when the laser beam strikes the road surface. Points representing obstacles are obtained when the laser beam strikes the obstacle. Therefore, if the points representing the road surface can be removed from the point cloud data, the position of the obstacles can be estimated from the remaining points. Obstacles refer to objects other than the road surface. Objects are things that reflect laser light and include living things such as people. Depending on the context, points representing the road surface may be called road surface reflection points, and points representing obstacles may be called obstacle points.

[0017] When the control device 21 acquires point cloud data from the sensor 30, it determines whether a combination of points aligned in the direction of travel of the vehicle 10 is a road surface reflection point or an obstacle point. If the sensor 30 changes its illumination angle in the horizontal direction, the point cloud data will contain multiple combinations of points aligned in the direction of travel of the vehicle 10. In this case, the control device 21 only needs to determine whether each combination of points aligned in the direction of travel of the vehicle 10 is a road surface reflection point or an obstacle point. The multiple points aligned in the direction of travel of the vehicle 10 may be points aligned in a straight line in the Y-axis direction. Also, depending on the scanning method of the sensor 30, the points may not be aligned in a straight line. In this case, the multiple points aligned in the direction of travel of the vehicle 10 may be slightly offset in the X-axis direction.

[0018] <Processing performed by the control device> The processing performed by the control device 21 when determining whether a point is a road surface reflection point or an obstacle point will be explained. As an example of a combination of multiple points aligned in the direction of travel of the vehicle 10, points P1 to P4 shown in Figure 1 will be used for the explanation.

[0019] As shown in Figure 2, in step S1, the control device 21 sets a virtual point PI. The virtual point PI is the point obtained by projecting the sensor 30 vertically onto the road surface. The control device 21 sets the virtual point PI from the mounting height of the sensor 30. The mounting height of the sensor 30 is a known value. The control device 21 shifts the coordinates of the sensor 30 in the Z-axis direction by the mounting height of the sensor 30. This gives the virtual point PI obtained by projecting the sensor 30 onto the road surface. The mounting height of the sensor 30 is stored in a storage device that the control device 21 can read. This storage device may be the storage unit 23.

[0020] Next, in step S2, the control device 21 determines whether the point is a road surface reflection point or not. When step S2 is performed for the first time, the point to be determined to be a road surface reflection point is the point closest to the sensor 30. Therefore, in the example shown in Figure 1, the determination of whether point P1 is a road surface reflection point or not is made. The control device 21 estimates the height of the point from the irradiation angle of the laser beam and the distance to the detected point. This height can also be called the Z coordinate of the point. The control device 21 compares the estimated height of the point with the mounting height of the sensor 30. The estimated height of the point is the height relative to the sensor 30. Therefore, when comparing the estimated height of the point with the mounting height of the sensor 30, the comparison may be made using absolute values. As a result of the comparison, if the difference between the estimated height of the point and the mounting height of the sensor 30 is less than the reference value, the control device 21 determines that the point is a road surface reflection point. If the difference between the estimated height of the point and the mounting height of the sensor 30 is greater than or equal to the reference value, the control device 21 determines that the point is an obstacle point. The reference value should be set according to the error that may occur in the estimated height of the point, based on the accuracy of the sensor 30, etc. If the judgment result of step S2 is negative, the control device 21 proceeds to step S3. If the judgment result of step S2 is positive, the control device 21 proceeds to step S4.

[0021] In step S3, the control device 21 moves to the next point. The next point is the point furthest from the sensor 30 after the point for which it was determined in step S2 whether or not it is a road surface reflection point. If it was determined in step S2 whether or not point P1 is a road surface, then in step S3 it moves to point P2. After completing the processing in step S3, the control device 21 returns to step S2. Then, in step S2, it is determined whether or not the points moved to in step S3 are road surface reflection points. That is, the determination in step S2 is repeated until one point is determined to be a road surface reflection point. In this way, the control device 21 makes determinations by comparing the mounting height of the sensor 30 with the height of the point, sequentially from the point closest to the sensor 30 in the direction away from the sensor 30, until it is determined that a point is a road surface reflection point by making a determination by comparing the mounting height of the sensor 30 with the height of the point.

[0022] In step S4, the control device 21 sets the following point as a road surface determination point. The following point is the point furthest from the sensor 30 after the point for which it was determined in step S2 whether or not it is a road surface reflection point. If it was determined in step S2 whether or not point P1 is a road surface, then point P2 is set as a road surface determination point. In this way, if the control device 21 determines that a point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point, it sets the point furthest from the sensor 30 after that point as a road surface determination point.

[0023] Next, in step S5, the control device 21 derives a straight line passing through two detection points that are located closer to the sensor 30 than the road surface determination point and have been determined to be detection points representing the road surface. The two detection points used to derive the straight line are preferably as close as possible to the road surface determination point. That is, the two detection points used to derive the straight line are preferably the detection point closest to the road surface determination point and the detection point second closest to the road surface determination point. The detection points are the virtual point PI and a plurality of points aligned in the direction of travel of the vehicle 10. In the example shown in Figure 1, the virtual point PI and points P1 to P4 are the detection points. The two detection points located closer to the sensor 30 than the road surface determination point are the point between the virtual point PI and the road surface determination point and two of the plurality of points including the virtual point PI. The detection point representing the road surface is the point determined to be a road surface reflection point in step S2 or step S8, or the virtual point PI.

[0024] If point P1 is determined to be a road surface reflection point in step S2, and point P2 is set as a road surface determination target point in step S4, the control device 21 derives a straight line passing through virtual point PI and point P1. If point P1 is determined to be an obstacle point in step S2, point P2 is determined to be a road surface reflection point, and point P3 is set as a road surface determination target point in step S4, the control device 21 derives a straight line passing through virtual point PI and point P2. Thus, the two detection points for deriving the straight line do not necessarily have to be adjacent detection points.

[0025] Next, in step S6, the control device 21 calculates the distance L1 from the road surface determination target point to the straight line derived in step S5. The distance L1 may be calculated using an approximation formula or using a map. As an example, the distance L1 can be obtained by L2 × sin(α + β). L2 is the distance between the detection point closest to the road surface determination target point and the road surface determination target point, of the two detection points from which the straight line was derived. α is the angle between the straight line passing through the two detection points and the horizontal plane. β is the angle between the straight line passing through the detection point closest to the road surface determination target point and the road surface determination target point and the horizontal plane. L2 and β can be calculated from the coordinates of the detection point closest to the road surface determination target point and the coordinates of the road surface determination target point, of the two detection points from which the straight line was derived. α can be derived from the coordinates of the two detection points from which the straight line was derived.

[0026] Figure 3 shows an example where point P2 is set as the road surface determination target point. In Figure 3, for the sake of explanation, the positional relationship between the virtual point PI and points P1 and P2 is exaggerated compared to Figure 1. The distance L1 is derived from point P2, which is the road surface determination target point, to the straight line L passing through the virtual point PI and point P1. As shown in Figure 3, the distance L1 in this embodiment is the distance in the direction perpendicular to the straight line L passing through the two detection points. The distance L1 may also be the distance in the direction perpendicular to the horizontal plane.

[0027] As described above, the control device 21 derives a straight line from the two detection points and the distance L1 from the road surface reflection point to the straight line. The straight line and distance L1 are derived in the YZ plane. Therefore, when deriving the straight line and distance L1, the control device 21 only needs to remove the X coordinate from the coordinates of the virtual point PI and points P1 to P4.

[0028] Next, in step S7, the control device 21 determines whether the distance L1 exceeds a threshold. The threshold is set so that it can determine whether the road surface determination target point is located on the same plane as the road surface represented by a straight line passing through the two detection points. Errors occur in the coordinates of the points based on the accuracy of the sensor 30, etc. The threshold should be set taking this error into consideration. In this embodiment, the threshold is a constant value. If the determination result in step S7 is negative, that is, if the distance L1 is less than or equal to the threshold, the control device 21 proceeds to step S8. If the determination result in step S7 is positive, the control device 21 proceeds to step S9.

[0029] In step S8, the control device 21 determines that the road surface determination target point is a road surface reflection point. In step S9, the control device 21 determines that the road surface determination target point is an obstacle point.

[0030] Next, in step S10, the control device 21 determines whether or not all points have been set as road surface determination points. All points mean all points that can be set as road surface determination points. That is, all points mean the point that was initially set as a road surface determination point in step S4, and points that are further from the sensor 30 than this point. If the determination result in step S10 is negative, the control device 21 returns to step S4. In this case, in step S4, the next point that is further from the sensor 30 after the point for which it was determined in step S7 whether or not it is a road surface reflection point should be set as a road surface determination point. That is, the control device 21 sequentially changes the road surface determination points from points close to the sensor 30 to points further away from the sensor 30. In the example shown in Figure 1, the road surface determination points are changed from point P2 → point P3 → point P4.

[0031] If point P3 is set as a road surface determination point, whether point P3 is a road surface reflection point is determined using a straight line passing through two points among the virtual point PI, point P1, and point P2 that are determined to represent the road surface. If points P1 and P2 are road surface reflection points, whether point P3 is a road surface reflection point is determined using a straight line passing through points P1 and P2. If point P1 is an obstacle point and point P2 is a road surface reflection point, whether point P3 is a road surface reflection point is determined using a straight line passing through virtual point PI and point P2.

[0032] If point P4 is set as a road surface determination point, whether point P4 is a road surface reflection point is determined using a straight line passing through two points from among the virtual point PI, points P1, P2, and P3 that are determined to represent the road surface. If points P2 and P3 are road surface reflection points, whether point P4 is a road surface reflection point is determined using a straight line passing through points P2 and P3. If point P2 is an obstacle point and points P1 and P3 are road surface reflection points, whether point P4 is a road surface reflection point is determined using a straight line passing through points P1 and P3. If point P3 is an obstacle point and points P1 and P2 are road surface reflection points, whether point P4 is a road surface reflection point is determined using a straight line passing through points P1 and P2.

[0033] If the result of step S10 is affirmative, the control device 21 terminates the process. [Operation of this embodiment] As shown in Figure 4, an obstacle O1 is present on the road surface. Assume that the point cloud data obtained in this state includes points P11, P12, and P13. If points P11 and P12 are determined to be road surface reflection points, a determination is made as to whether point P13 is a road surface reflection point or not using a straight line passing through points P11 and P12.

[0034] Since the obstacle O1 is located on the road surface, the obstacle point can be obtained as a coordinate at a position higher than the road surface. As a result, the distance L1 between the line passing through point P13, which is the road surface target, and points P11 and P12, which are road surface reflection points, becomes longer. This allows the control device 21 to determine that point P13 is an obstacle point.

[0035] The control device 21 may estimate the position of an obstacle point as the position of an obstacle. Alternatively, the control device 21 may perform clustering, which combines multiple obstacle points into one, and treat the clustered points as an obstacle. In this case, the position of the obstacle may be the coordinates of a representative point among the clustered points. In this way, the position estimation device 20 estimates the position of the obstacle.

[0036] [Effects of this embodiment] (1) A straight line passing through two detection points represents the slope of the road surface. Therefore, if the distance L1 is less than or equal to a threshold, the point to be determined to be a road surface can be considered to represent the road surface. By deriving a straight line passing through two detection points for each point to be determined to be a road surface, the slope of the vicinity can be derived for each point to be determined to be a road surface. Therefore, even if the slope of the road surface is not uniform, it is possible to determine whether a point to be determined to be a road surface is a point that represents the road surface by comparing it with the slope of the vicinity. Therefore, even if the slope of the road surface is not uniform, it is possible to determine whether a point represents the road surface or not.

[0037] (2) When the control device 21 determines whether the point closest to the sensor 30 is a road surface reflection point, it estimates the height of the point from the irradiation angle of the laser beam and the distance to the detected point. The control device 21 then determines whether the point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point. If the point closest to the sensor 30 is not a road surface reflection point, the control device 21 determines whether the next point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point. The control device 21 then repeats this determination until one point is determined to be a road surface reflection point.

[0038] To derive a straight line, two detection points representing the road surface are required. Since the virtual point PI is a detection point representing the road surface, one more point representing the road surface is needed to derive a straight line. Therefore, whether a point is a road surface reflection point or not is determined by comparing the mounting height of the sensor 30 with the height of the point.

[0039] When determining whether a point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point, the height of the point calculated due to the difference between the assumed illumination angle and the actual illumination angle may contain errors depending on the mounting accuracy of the sensor 30. The error becomes larger the further the point is from the sensor 30. For this reason, for points close to the sensor 30, even when determining whether a point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point, the influence of mounting accuracy is small. For points close to the sensor 30, determining whether a point is a road surface reflection point by comparing the mounting height of the sensor 30 with the height of the point does not require excessively high mounting accuracy of the sensor 30. Therefore, the process of adjusting the mounting position of the sensor 30 can be reduced.

[0040] (3) The detection point includes a virtual point PI. The virtual point PI is the point obtained by projecting the sensor 30 onto the road surface, and can therefore be said to represent the road surface. By using the virtual point PI as a detection point, if it can be determined that one of the points is a road surface reflection point, a straight line can be derived.

[0041] (4) One method for determining whether a point is a road surface reflection point is to consider two points as obstacle points if the distance between two consecutive points in the vertical direction is short. In this method, two or more points in the vertical direction are required to detect an obstacle. For this reason, thin obstacles may not be detected as obstacles. In particular, the further away from the sensor 30 the point density decreases, so obstacles that are farther away become more difficult to detect. In contrast, with this embodiment, an obstacle can be detected even if only one point is present. For this reason, even thin obstacles or obstacles that are far away can be detected.

[0042] [Examples of Modifications] The embodiment can be implemented with the following modifications. The embodiment and the following examples of modifications can be combined with each other to the extent that they do not contradict each other technically.

[0043] ○When determining whether the point closest to the sensor 30 is a road surface reflection point, the control device 21 may determine whether it is a road surface reflection point by a method different from the embodiment. For example, since obstacle points are points arranged vertically, it can be estimated that points arranged horizontally are road surface reflection points. Therefore, it is also possible to determine whether the point closest to the sensor 30 is a road surface reflection point by comparing it with the second-closest point to the sensor 30. For example, if the point closest to the sensor 30 and the second-closest point to the sensor 30 are not arranged vertically, the control device 21 may determine that the point closest to the sensor 30 is a road surface reflection point.

[0044] ○The control device 21 may regard the point closest to the sensor 30 as a road surface reflection point without determining whether it is a road surface reflection point. Since the vehicle 10 moves, the distance between the obstacle and the vehicle 10 decreases as the vehicle 10 moves. Therefore, the closer the obstacle is to the vehicle 10, the higher the possibility that it has been detected as an obstacle in the process of the vehicle 10 moving. The vehicle 10 moves away from obstacles that hinder progress by an avoidance operation, so the possibility that an obstacle that hinders progress is near the vehicle 10 is low. Therefore, the point closest to the sensor 30 is unlikely to be an obstacle point and can be regarded as a road surface reflection point.

[0045] ○The control device 21 may increase the threshold value as the distance from the sensor 30 increases. The error in the point coordinates becomes larger at positions farther from the sensor 30. Therefore, by increasing the threshold value, the influence of the error can be reduced.

[0046] ○The position estimation device 20 may estimate the self-position of the vehicle 10. In this case, the control device 21 estimates the self-position by comparing the obstacle points with a map representing the environment. ○The sensor 30 may be provided so as to irradiate laser light backward. In this case, the position estimation device 20 can estimate the position of an obstacle behind the vehicle 10.

[0047] ○ The sensor 30 only needs to be able to emit laser light at predetermined angles relative to the vertical direction, and does not need to change the irradiation angle of the laser light in the horizontal direction. ○ The detection point does not need to include the virtual point PI. In this case, the control device 21 can extract two road surface reflection points by comparing the mounting height of the sensor 30 with the height of the estimated point.

Claims

1. A position estimation device for a vehicle, comprising: a sensor configured to detect the distance to a point struck by electromagnetic waves by irradiating them at predetermined angles with respect to the vertical; and a control device, wherein a plurality of the points aligned in the direction of travel of the vehicle are detection points, and the control device sets one of the plurality of points aligned in the direction of travel, excluding the point closest to the sensor, as a road surface determination target point, determines whether the detection point located closer to the sensor than the road surface determination target point represents a point on the road surface, derives a straight line passing through the two detection points that are located closer to the sensor than the road surface determination target point and have been determined to represent the road surface, determines that the road surface determination target point represents the road surface if the distance between the straight line and the road surface determination target point is less than or equal to a threshold, and sequentially changes the road surface determination target point from the point closest to the sensor to the point further away from the sensor.

2. The control device is further configured to estimate the height of the point from the irradiation angle of the electromagnetic wave and the distance to the detected point, to determine whether the point represents the road surface by comparing the mounting height of the sensor with the height of the point, to perform the comparison of the mounting height of the sensor with the height of the point sequentially from the point closest to the sensor in the direction away from the sensor until it is determined that the point represents the road surface by the comparison of the mounting height of the sensor with the height of the point, and if it is determined that the point represents the road surface by the comparison of the mounting height of the sensor with the height of the point, to set the next point farther from the sensor after that point as the road surface determination target point.

3. The position estimation device according to claim 1, wherein the detection point includes a virtual point projected onto the road surface by the sensor.

4. The position estimation device according to claim 1, wherein the control device increases the threshold as it moves away from the sensor.

5. A position estimation method applicable to a vehicle, comprising: irradiating a sensor with electromagnetic waves at predetermined angles with respect to the vertical and detecting the distance to the point struck by the electromagnetic waves, wherein a plurality of the points aligned in the direction of travel of the vehicle are detection points, and the position estimation method further comprises: setting one of the plurality of points aligned in the direction of travel, excluding the point closest to the sensor, as a road surface determination target point; determining whether the detection point located closer to the sensor than the road surface determination target point represents a point on the road surface; deriving a straight line passing through two detection points that are located closer to the sensor than the road surface determination target point and have been determined to represent the road surface; determining that the road surface determination target point represents the road surface if the distance between the straight line and the road surface determination target point is less than or equal to a threshold; and sequentially changing the road surface determination target points from the point closest to the sensor to the point further away from the sensor.

Images