Vehicle position estimation device, vehicle control system, and vehicle position estimation method

Abstract

To obtain an own vehicle position estimation device capable of continuously performing highly accurate own vehicle position estimation.SOLUTION: An own vehicle position estimation device comprises: a first road section line information acquisition section 101 that acquires first road section line information including first left-side road section line information and first right-side road section line information detected by an imaging element mounted on a vehicle 10; a second road section line information acquisition section 102 that acquires second road section line information including second left-side road section line information and second right-side road section line information based on a locator 91 and map information; a road section line information determination section 103 that determines a combination of road section line information to be used for estimating an own vehicle position on the basis of the first road section line information and the second road section line information; and an own vehicle position estimation section 104 that estimates the own vehicle position by correcting the second road section line information using the combination of the road section line information determined by the road section line information determination section 103.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present application relates to a vehicle position estimation device, a vehicle control system, and a vehicle position estimation method. [Background technology]

[0002] In recent years, with the aim of realizing autonomous vehicle driving, technologies have been developed that estimate the vehicle's position on a map with high accuracy using road-dividing line information acquired by sensors such as cameras mounted on the vehicle and road-dividing line information from high-precision maps.

[0003] However, the recognition of road dividing lines by a camera has a problem in that road dividing lines may not be recognized with the accuracy required for autonomous driving due to factors such as the disappearance, blurring, or omission of white lines on the road. Furthermore, camera-related problems such as camera lens dirt, flare, and occlusion may prevent road dividing lines from being recognized with the accuracy required for autonomous driving, or may result in the incorrect recognition of objects that are not road dividing lines as road dividing lines. In other words, the above-described vehicle position estimation technology has a problem in that if the camera incorrectly recognizes road dividing lines, the vehicle position cannot be correctly estimated. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7184951 Summary of the Invention [Problem to be solved by the invention]

[0005] The vehicle control device and vehicle control method described in Patent Document 1 disclose a vehicle control technology that includes a judgment unit that compares two road dividing lines obtained by two different means to determine whether the road dividing line information of either one has been erroneously recognized.

[0006] The vehicle control device and vehicle control method described in Patent Document 1 disclose a method for determining whether the recognition result of a road lane marking is erroneous. However, the device and method do not take into consideration a configuration and method for continuing vehicle control processing when an erroneous recognition of a road lane marking occurs.

[0007] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a vehicle position estimation device, a vehicle control system, and a vehicle position estimation method that are capable of continuously estimating the vehicle position with high accuracy by dynamically changing the road dividing line information used for estimating the vehicle position to obtain the optimal combination of information. [Means for solving the problem]

[0008] The vehicle position estimation device disclosed in the present application comprises: A vehicle position estimation device that estimates the position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, a first road-dividing line information acquisition unit that acquires first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; a second road-dividing line information acquisition unit that acquires second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; a road-dividing line information determination unit that determines a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; a vehicle position estimation unit that estimates the vehicle position by correcting the second road-dividing line information using a combination of the road-dividing line information determined by the road-dividing line information determination unit; and Equipped with.

[0009] The vehicle control system disclosed in the present application comprises: a vehicle position estimation device that estimates a vehicle position based on the road division line information; a driving route generation device that generates a driving route for the vehicle to reach a target point based on the vehicle position output from the vehicle position estimation device; and a vehicle control device that sets a target trajectory and a target vehicle speed to be used when performing vehicle control of the vehicle on the generated travel route.

[0010] The vehicle position estimation method disclosed in the present application includes: A vehicle position estimation method for estimating a vehicle position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, using a vehicle position estimation device, comprising: acquiring first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; acquiring second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; determining a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; and correcting the second road-dividing line information using the determined combination of road-dividing line information to estimate the vehicle position. [Effects of the Invention]

[0011] According to the vehicle position estimation device, vehicle control system, and vehicle position estimation method disclosed in the present application, the road dividing line information used for vehicle position estimation is dynamically changed to obtain the optimal combination of information, thereby achieving the effect of continuously enabling highly accurate vehicle position estimation. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram illustrating a configuration of a vehicle position estimation device according to a first embodiment. [Figure 2]1 is a schematic diagram of a vehicle equipped with a vehicle position estimation device according to a first embodiment. [Figure 3] 1 is a schematic diagram showing a situation in which a vehicle equipped with a vehicle position estimation device according to a first embodiment is traveling in a traffic lane. [Figure 4] 3 is a schematic diagram for explaining the operation of a first road dividing line information acquisition unit that is part of the configuration of the vehicle position estimation device according to the first embodiment. FIG. [Figure 5] 2 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the first embodiment. FIG. [Figure 6] 2 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the first embodiment. FIG. [Figure 7] 2 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the first embodiment. FIG. [Figure 8] 2 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the first embodiment. FIG. [Figure 9] 2 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the first embodiment. FIG. [Figure 10] FIG. 2 is a flowchart illustrating a vehicle position estimation method according to the first embodiment. [Figure 11] FIG. 10 is a block diagram illustrating a configuration of a vehicle position estimation device according to a second embodiment. [Figure 12] FIG. 10 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the second embodiment. [Figure 13] FIG. 10 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the second embodiment. [Figure 14] FIG. 10 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the second embodiment. [Figure 15] FIG. 10 is a schematic diagram for explaining the operation of the vehicle position estimation device according to the second embodiment. [Figure 16] FIG. 10 is a flowchart illustrating a vehicle position estimation method according to the second embodiment. [Figure 17] FIG. 10 is a block diagram illustrating a configuration of a vehicle control system according to a fourth embodiment. [Figure 18] 1 is a block diagram showing a hardware configuration for realizing a vehicle position estimation device according to the first and second embodiments and a vehicle control system according to the fourth embodiment. [Figure 19] 1 is a block diagram showing a hardware configuration for realizing a vehicle position estimation device according to the first and second embodiments and a vehicle control system according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Embodiment 1 <Configuration of Vehicle Position Estimation Device According to First Embodiment> 1 is a block diagram showing the configuration of a vehicle position estimation device 100 according to the first embodiment. The vehicle position estimation device 100 according to the first embodiment includes a first road-dividing line information acquisition unit 101, a second road-dividing line information acquisition unit 102, a road-dividing line information determination unit 103, and a vehicle position estimation unit 104. The vehicle position estimation device 100 outputs the vehicle position estimation result to a driving route generation device 200, and the driving route generation device 200 outputs a driving route generated based on the vehicle position estimation result to a vehicle control device 300.

[0014] A camera 90 mounted on the vehicle 10 outputs an image of the road conditions ahead of the vehicle 10 to a first road-dividing line information acquisition unit 101. A locator 91 mounted on the vehicle 10 outputs the acquired vehicle position information to a second road-dividing line information acquisition unit 102.

[0015] 2, the vehicle position estimation device 100, the driving route generation device 200, and the vehicle control device 300 according to the first embodiment are incorporated into the vehicle 10. The driving route generation device 200 generates a driving route using the vehicle position estimation result output from the vehicle position estimation device 100, and the vehicle control device 300 generates target control amounts, such as a target steering amount and a target acceleration / deceleration, which are required for automatic driving control of the vehicle 10 on the generated driving route, and outputs the target control amounts to an actuator 530 installed in the vehicle 10, thereby performing vehicle control of the vehicle 10.

[0016] The first road-dividing line information acquisition unit 101 captures images of the area around the vehicle 10 using the camera 90, and acquires and outputs position information of the left and right road-dividing lines of the lane 20 in which the vehicle 10 is traveling. Hereinafter, position information of the road-dividing line on the left side in the direction of travel of the vehicle 10 captured by the camera 90 will be referred to as first left-side road-dividing line information, and position information of the road-dividing line on the right side in the direction of travel of the vehicle 10 will be referred to as first right-side road-dividing line information. Furthermore, the first left-side road-dividing line information and the first right-side road-dividing line information will be collectively referred to as first road-dividing line information.

[0017] 3 is a schematic diagram showing the state of a lane 20 when a vehicle 10 is traveling along it. Generally, white lines are provided on lanes as road dividing lines. On the left side of the lane 20 in the direction of travel of the vehicle 10, a first left-hand road dividing line 21, represented by a white line, for example, is provided. On the right side of the lane 20 in the direction of travel of the vehicle 10, a first right-hand road dividing line 22, represented by a white line, for example, is provided.

[0018] Fig. 4 is a schematic diagram for explaining the operation of the first road-dividing line information acquisition unit 101. The diagram on the left side of Fig. 4 is a schematic diagram of an image of the lane 20 captured by the camera 90 mounted on the vehicle 10. On the left side of the vehicle 10, a white line that serves as the basis for the first left-side road-dividing line information is captured. On the right side of the vehicle 10, a white line that serves as the basis for the first right-side road-dividing line information is captured.

[0019] The first road-dividing line information acquisition unit 101 converts the white lines provided as road-dividing lines on the left and right sides of the lane 20 into first point cloud data 25. The first point cloud data 25 corresponding to the first left-side road-dividing line 21 is referred to as first left-side point cloud data 25a, and the first point cloud data 25 corresponding to the first right-side road-dividing line 22 is referred to as first right-side point cloud data 25b. In other words, the first left-side road-dividing line information is made up of the points of the first left-side point cloud data 25a, and the first right-side road-dividing line information is made up of the points of the first right-side point cloud data 25b. The diagram on the right side of FIG. 4 shows the state after the white lines have been converted into the first point cloud data 25. Note that point cloud data refers to a collection of points indicating the positions of road-dividing lines at regular intervals in the longitudinal direction of the vehicle 10.

[0020] The second road-dividing line information acquisition unit 102 stores map information and acquires and outputs position information of the left and right road-dividing lines of the lane 20 in which the vehicle 10 is traveling, based on the vehicle position information acquired by the locator 91 and this map information. Hereinafter, the position information of the road-dividing line on the left side in the direction of travel of the vehicle 10, based on the vehicle position information and map information, will be referred to as second left-side road-dividing line information, and the position information of the road-dividing line on the right side in the direction of travel of the vehicle 10, will be referred to as second right-side road-dividing line information. Furthermore, the second left-side road-dividing line information and the second right-side road-dividing line information will be collectively referred to as second road-dividing line information. Figure 5 is a schematic diagram showing the state of the lane 20 on a map. A second left-side road-dividing line 23 based on the map information is displayed on the left side of the lane 20, and a second right-side road-dividing line 24 based on the map information is displayed on the right side of the lane 20.

[0021] The second road-dividing line information acquisition unit 102 converts the left and right road-dividing lines of the lane 20 into second point cloud data 26 based on the vehicle position information and map information acquired by the locator 91. The second point cloud data 26 corresponding to the second left-side road-dividing line 23 is referred to as second left-side point cloud data 26a, and the second point cloud data 26 corresponding to the second right-side road-dividing line 24 is referred to as second right-side point cloud data 26b. In other words, the second left-side road-dividing line information is made up of the points of the second left-side point cloud data 26a, and the second right-side road-dividing line information is made up of the points of the second right-side point cloud data 26b. FIG. 6 shows the state after the road-dividing lines based on the map information have been converted into second point cloud data 26.

[0022] Here, point cloud data will be described using FIG. 7. FIG. 7 is a schematic diagram illustrating point cloud data extracted from road-dividing line information. For example, if the road-dividing line information is based on white lines, the road-dividing line on the left side of the vehicle 10 in the traveling direction is represented by the points of the first left-side point cloud data 25a, which is the first left-side road-dividing line information. In FIG. 7, the first left-side point cloud data 25a is composed of point cloud data consisting of a total of six points. That is, the white line on the left side of the vehicle 10 in the traveling direction is represented by the first left-side point cloud data 25a, which is point cloud data consisting of six points extracted from the white line. That is, the first left-side point cloud data 25a is the first left-side road-dividing line information acquired while the vehicle 10 is traveling. In the example shown in FIG. 7, the interval between the individual points of the first left-side point cloud data 25a is set to 10 m. However, the interval between the point cloud data is not limited to 10 m and may be set appropriately depending on the road conditions, the vehicle speed of the vehicle 10, etc.

[0023] In the first road-dividing line information and the second road-dividing line information, the spacing between points in the respective point cloud data in the longitudinal direction of the vehicle 10 is approximately the same. However, the ranges that can be represented by the point cloud data in the first road-dividing line information and the second road-dividing line information do not necessarily match. For example, the first road-dividing line information may be set within a range from 100 m ahead to 50 m behind the vehicle 10, while the second road-dividing line information may be set within a range from 300 m ahead to 100 m behind the vehicle 10. Note that this setting method is merely an example, and the settings may be made appropriately depending on the road conditions, the vehicle speed of the vehicle 10, etc.

[0024] The road-dividing line information determination unit 103 selects road-dividing line information that provides the most optimal vehicle position estimation result from among the vehicle position estimation using the left road-dividing line, the vehicle position estimation using the right road-dividing line, and the vehicle position estimation using the left and right road-dividing lines, based on, for example, first left-side road-dividing line information made up of first left-side point cloud data 25a extracted from white lines on the left side of lane 20 and first right-side road-dividing line information made up of first right-side point cloud data 25b extracted from white lines on the right side of lane 20 (i.e., first road-dividing line information made up of first point cloud data 25), and second left-side road-dividing line information made up of second left-side point cloud data 26a representing the left road-dividing line of the lane and second right-side road-dividing line information made up of second right-side point cloud data 26b representing the right road-dividing line of lane 20, both of which are acquired from the vehicle position information and map information acquired by locator 91. The lane-dividing line information determining unit 103 outputs the selected lane-dividing line information as best lane-dividing line information.

[0025] On the other hand, if neither the left-side road-dividing line information nor the right-side road-dividing line information can be used to estimate the vehicle's position, the road-dividing line information determination unit 103 determines that the vehicle's position cannot be estimated and outputs "invalid" as the best road-dividing line information.

[0026] If the best road-dividing line information output by the road-dividing line information determination unit 103 is valid, the vehicle position estimation unit 104 estimates the vehicle position by appropriately correcting the position of the second road-dividing line information using road-dividing line information on at least one of the left and right sides from the first road-dividing line information and the second road-dividing line information in accordance with the best road-dividing line information. The above is an outline of the configuration of the vehicle position estimation device 100 according to the first embodiment.

[0027] Next, the camera 90 and the locator 91 connected to the vehicle position estimation device 100 will be described below. The camera 90 is installed in front of the vehicle 10 at a position where it can detect road dividing lines, such as white lines, as an image. The camera 90 outputs the captured image to the first road dividing line information acquisition unit 101. Note that the camera 90 is merely one example of an image sensor, and any image sensor that can detect road dividing lines as an image can be used, and is not limited to the camera 90.

[0028] Locator 91 is a device for acquiring position information of vehicle 10, and is configured, for example, with a GPS (Global Positioning System) receiver. Alternatively, the vehicle position may be measured using a GNSS (Global Navigation Satellite System) receiver. Furthermore, the vehicle position may be detected by a device such as an inertial navigation system that uses various sensors mounted on vehicle 10. Locator 91 outputs the acquired vehicle position information to second road dividing line information acquisition unit 102.

[0029] 1, a vehicle position estimation device 100 estimates the vehicle position of a vehicle 10 and outputs the vehicle position estimation result to a driving route generation device 200. The driving route generation device 200 generates a driving route based on the vehicle position estimation result and outputs the generated driving route to a vehicle control device 300. The vehicle control device 300 optimally controls vehicle control, such as automatic driving, of the vehicle 10 based on the generated driving route.

[0030] The operations of the lane dividing line information determining unit 103 and the vehicle position estimating unit 104, which are characteristic of the configuration of the vehicle position estimating device 100 according to the first embodiment, will be described in more detail below.

[0031] <Operation of the road lane marking information determination unit 103> The lane dividing line information determining unit 103 performs the following processing. (1) Regarding the left road-dividing line, a left deviation LDv is calculated, which indicates the deviation between the first left road-dividing line information and the second left road-dividing line information. The method for calculating the deviation is described below. (2) With respect to the first left-side point cloud data 25a, which is the first left-side road dividing line information, and the second left-side point cloud data 26a, which is the second left-side road dividing line information, a process is performed for all point cloud data included within the distance range expressed by both the first left-side road dividing line information and the second left-side road dividing line information to calculate the amount of deviation Δd in the left-right direction of the vehicle 10 between points in the point cloud data in the fore-and-aft direction of the vehicle 10 where the inter-point distance between the points is the same. (3) The deviation Δd in the left-right direction of the vehicle 10 is calculated between a point at a distance of 0 m in the fore-and-aft direction of the vehicle 10 in the first left-side road-dividing line information and a point at a distance of 0 m in the fore-and-aft direction of the vehicle 10 in the second left-side road-dividing line information. In other words, the deviation Δd at a distance of 0 m in the fore-and-aft direction of the vehicle 10 is calculated. Similarly, the deviation Δd in the left-and-right direction is calculated sequentially for points at distances of 10 m, 20 m, and within the same distance range in both directions. (4) The sum Δdsum of the absolute values ​​of the deviation amounts Δd in the left-right direction of the vehicle 10 is set as the left deviation degree LDv of the left road dividing line information.

[0032] By calculating the left deviation degree LDv as described above, it is possible to quantify the degree of deviation between the left-side road-dividing line information represented by the first left-side road-dividing line information and the left-side road-dividing line information represented by the second left-side road-dividing line information. In the following description, the amount of deviation in the left-right direction is simply referred to as the deviation amount.

[0033] An example of a method for calculating the left-side deviation degree LDv is shown in Figure 8. The left-hand diagram in Figure 8 shows first left-side road-dividing line information represented by first left-side point cloud data 25a consisting of four points extracted from an image acquired by capturing the first left-side road-dividing line, which is represented by a white line, with camera 90. The center diagram in Figure 8 shows second left-side road-dividing line information represented by second left-side point cloud data 26a consisting of six points extracted from the second left-side road-dividing line based on the vehicle position information and map information acquired by locator 91.

[0034] As shown in the diagram on the right side of Figure 8, points between the first left-side point cloud data 25a, which is the first left-side road-dividing line information, and the second left-side point cloud data 26a, which is the second left-side road-dividing line information, at which the distance in the fore-and-aft direction of the vehicle 10 matches, are plotted on the same plane, and the distance between the two, i.e., the deviation amount Δd, is calculated for each point.

[0035] In the example shown in Figure 8, when the distance L in the fore-and-aft direction of the vehicle 10 is -10 m, the deviation amount Δd is calculated as -30 cm; when the distance L is 0 m, the deviation amount Δd is +5 cm; when the distance L is 10 m, the deviation amount Δd is +50 cm; and when the distance L is 20 m, the deviation amount Δd is +100 cm.

[0036] The sum Δdsum of the absolute values ​​of the deviation Δd is the sum of the absolute values ​​of the deviation Δd at all points. In the example shown in Figure 8, the sum Δdsum of the absolute values ​​of the deviation Δd is 185 cm. The left deviation LDv is expressed by Δdsum.

[0037] The left-side deviation LDv may be quantified using a method other than the sum Δdsum of the absolute values ​​of the deviations Δd. The left-side deviation LDv may be calculated by using the maximum absolute value of the deviations Δd at each point of the point cloud data, or by calculating the mean square error of the deviations Δd.

[0038] In addition to the above-described method of calculating the left deviation degree LDv, for example, before calculating the deviation amount Δd, the second point cloud data 26 may be translated so that the center of gravity positions of both the first point cloud data 25, which is the first road-dividing line information, and the second point cloud data 26, which is the second road-dividing line information, coincide for all points included within the range of the longitudinal distance of the vehicle 10 represented by both the first road-dividing line information and the second road-dividing line information. Applying such translation processing makes it possible to calculate the left deviation degree LDv with emphasis on the deviation amount in the lateral direction of the road-dividing line information.

[0039] Fig. 9 shows an example of calculating the deviation degree taking into account the position of the center of gravity of the point cloud data. The left diagram in Fig. 9 shows first left-side road-dividing line information represented by first left-side point cloud data 25a consisting of four points extracted from an image acquired by capturing the first left-side road-dividing line, represented by a white line, with camera 90, and the position of center of gravity 30a of the first left-side point cloud data 25a consisting of four points. The center diagram in Fig. 9 shows second left-side road-dividing line information represented by second left-side point cloud data 26a consisting of six points extracted from the second left-side road-dividing line based on map information, and the position of center of gravity 31a of the second left-side point cloud data 26a consisting of six points.

[0040] As shown in the right diagram in Figure 9, the second left-side point cloud data 26a is translated so that the center of gravity position 30a of the first left-side point cloud data 25a, which is the first left-side road-dividing line information, and the center of gravity position 31a of the second left-side point cloud data 26a, which is the second left-side road-dividing line information, coincide with each other.Furthermore, points that are the same distance in the fore-and-aft direction of the vehicle 10 are plotted on the same plane, and the distance between the two, i.e., the deviation amount Δd, is calculated for each point.

[0041] In the example shown in Figure 9, when the distance L in the fore-and-aft direction of the vehicle 10 is -10 m, the deviation amount Δd is calculated as -50 cm; when the distance L is 0 m, the deviation amount Δd is -25 cm; when the distance L is 10 m, the deviation amount Δd is +25 cm; and when the distance L is 20 m, the deviation amount Δd is +50 cm.

[0042] 9, the sum Δdsum of the absolute values ​​of the deviation amounts Δd is 150 cm. That is, the left deviation LDv is 150 cm.

[0043] For the right-side road-dividing line information, the right deviation degree RDv is calculated using the same method as for the left-side road-dividing line information.

[0044] The road-dividing line information determination unit 103 determines whether the second left-side road-dividing line information is erroneously recognized relative to the first left-side road-dividing line. Specifically, if the left deviation degree LDv is equal to or greater than a preset threshold value Dth, the road-dividing line information determination unit 103 determines that the second left-side road-dividing line information is erroneously recognized. On the other hand, if the left deviation degree LDv is less than the preset threshold value Dth, the road-dividing line information determination unit 103 determines that the second left-side road-dividing line information is not erroneously recognized.

[0045] The road-dividing line information determination unit 103 further determines whether the second right-side road-dividing line information is erroneously recognized relative to the first right-side road-dividing line. Specifically, if the right deviation degree RDv is equal to or greater than a preset threshold value Dth, the road-dividing line information determination unit 103 determines that the second right-side road-dividing line information is erroneously recognized. On the other hand, if the right deviation degree RDv is less than the preset threshold value Dth, the road-dividing line information determination unit 103 determines that the second right-side road-dividing line information is not erroneously recognized.

[0046] The lane dividing line information determining unit 103 outputs the following based on the above-mentioned determination result. (1) When there are no errors in the second left-side road dividing line information and the second right-side road dividing line information The lane-dividing line information determining unit 103 outputs both the second left-side lane-dividing line information and the second right-side lane-dividing line information as the best lane-dividing line information. (2) When there is no misrecognition in the second left road dividing line information, but there is a misrecognition in the second right road dividing line information The lane-dividing line information determining unit 103 outputs only the second left-side lane-dividing line information as the best lane-dividing line information. (3) When there is an error in the second left-side road dividing line information and there is no error in the second right-side road dividing line information The lane-dividing line information determining unit 103 outputs only the second right-side lane-dividing line information as the best lane-dividing line information. (4) When there is an error in both the second left road dividing line information and the second right road dividing line information The lane-dividing line information determining unit 103 outputs "invalid" as the best lane-dividing line information. The above is an outline of the operation of the lane dividing line information determining unit 103.

[0047] <Operation of the vehicle position estimation unit 104> The vehicle position estimation unit 104 targets several points of the second road-dividing line information near the vehicle 10 and calculates the rotation angle and translation amount of each point of the second road-dividing line information so that the points of the second road-dividing line information match the positions of the points of the first road-dividing line information as closely as possible.The vehicle position estimation unit 104 estimates the vehicle position by rotating and translating the entire second road-dividing line using the calculated rotation angle and translation amount.The calculated rotation angle refers to a correction angle, which is one of the correction amounts.

[0048] The vehicle position estimation unit 104 performs the following processing based on the best lane-dividing line information output as the determination result by the lane-dividing line information determination unit 103. (1) When the best road lane marking information is "both the second left road lane marking information and the second right road lane marking information" The vehicle position estimation unit 104 calculates the rotation angle and the translation amount, that is, the correction amount, based on both the second left-side road-dividing line information and the second right-side road-dividing line information. (2) When the best road lane marking information is "only the second left road lane marking information" The vehicle position estimation unit 104 calculates the rotation angle and the translation amount, that is, the correction amount, based only on the second left-side road dividing line information. (3) When the best road lane marking information is "only the second right road lane marking information" The vehicle position estimation unit 104 calculates the rotation angle and the translation amount, that is, the correction amount, based only on the second right-side road dividing line information. (4) When the best road lane marking information is "invalid" The vehicle position estimation unit 104 does not output the estimation result of the vehicle position.

[0049] If the best road dividing line information is "both the second left-side road dividing line information and the second right-side road dividing line information," the rotation angle and parallel movement amount will be calculated using twice the number of points compared to when only one of the information is present.

[0050] The reason for the above processing is as follows. The second road-dividing line information based on map information has the correct shape of the road-dividing line due to the accuracy of the map information, but because the vehicle position information acquired by locator 91 contains measurement errors, the map information based on the vehicle position information has the property that the position of the road-dividing line relative to vehicle 10 is inaccurate. On the other hand, the first road-dividing line information based on the image captured by camera 90 has the property that the position relative to vehicle 10 is accurate for points near vehicle 10, but measurement errors become larger for points far from vehicle 10. Therefore, by processing the first road-dividing line information and the second road-dividing line information as described above, the disadvantages of both are canceled out and the advantages of both are synergistically utilized, resulting in more accurate road-dividing line information.

[0051] <Method for estimating vehicle position according to first embodiment> A vehicle position estimation method using the vehicle position estimation device 100 according to the first embodiment will be described below with reference to the flowchart in Fig. 10. Note that the final step, step S112, corresponds to the operations of the driving route generation device 200 and the vehicle control device 300 based on the output of the vehicle position estimation device 100.

[0052] First, in step S101, the first road dividing line information acquisition unit 101 photographs the surroundings of the vehicle 10 using the camera 90, outputs first road dividing line information consisting of first left side road dividing line information and first right side road dividing line information, which are position information of the road dividing lines on the left and right sides of the lane 20 in which the vehicle 10 is traveling, and proceeds to processing in step S102.

[0053] In step S102, the second road dividing line information acquisition unit 102 outputs second road dividing line information consisting of second left side road dividing line information and second right side road dividing line information, which are position information of the road dividing lines on the left and right sides of the lane 20 in which the vehicle 10 is traveling, based on the vehicle position information and map information acquired by the locator 91, and proceeds to processing in step S103.

[0054] In step S103, the lane-dividing line information determining unit 103 calculates the deviation (left deviation LDv) between the first left-side lane-dividing line information and the second left-side lane-dividing line information for the left-side lane-dividing line, and then proceeds to the processing of step S104.

[0055] In step S104, the lane-dividing line information determining unit 103 calculates the deviation (right deviation RDv) between the first right-side lane-dividing line information and the second right-side lane-dividing line information for the right-side lane-dividing line, and then proceeds to the processing of step S105.

[0056] In step S105, the lane-dividing line information determination unit 103 determines whether the deviation of the left-side lane-dividing line information (left-side deviation LDv) is equal to or greater than the threshold Dth. If the deviation is equal to or greater than the threshold Dth, the process proceeds to step S106. If the deviation is less than the threshold Dth, the process proceeds to step S110.

[0057] In step S106, the road-dividing line information determination unit 103 determines whether the deviation degree of the right-side road-dividing line information (right-side deviation degree RDv) is equal to or greater than the threshold value Dth. If the deviation degree is equal to or greater than the threshold value Dth, this corresponds to the case where there is a recognition error in both the second left-side road-dividing line information and the second right-side road-dividing line information, so the road-dividing line information determination unit 103 outputs "invalid" as the best road-dividing line information and proceeds to the processing of step S107. On the other hand, if the deviation degree is less than the threshold value Dth, this corresponds to the case where there is a recognition error in the second left-side road-dividing line information but there is no recognition error in the second right-side road-dividing line information, so the road-dividing line information determination unit 103 outputs only the second right-side road-dividing line information as the best road-dividing line information and proceeds to the processing of step S108.

[0058] In step S107, based on the result that the road lane marking information determination unit 103 outputs "invalid" as the best road lane marking information, the vehicle position estimation unit 104 determines that it is not possible to estimate the vehicle position and does not output the estimation result of the vehicle position.

[0059] If it is determined in step S106 that the deviation of the right-side road dividing line information (right-side deviation RDv) is less than the threshold value Dth, in step S108, the vehicle position estimation unit 104 estimates the vehicle position using only the second right-side road dividing line information, which is the best road dividing line information, and proceeds to processing in step S112.

[0060] Next, the process in step S105 when the deviation is less than the threshold value Dth will be described. In step S110, the road-dividing line information determination unit 103 determines whether the deviation degree of the right-side road-dividing line information (right-side deviation degree RDv) is equal to or greater than the threshold value Dth. If the deviation degree is equal to or greater than the threshold value Dth, this corresponds to the case where there is no misrecognition in the second left-side road-dividing line information but there is a misrecognition in the second right-side road-dividing line information. Therefore, the road-dividing line information determination unit 103 outputs only the second left-side road-dividing line information as the best road-dividing line information and proceeds to the processing of step S109. On the other hand, if the deviation degree is less than the threshold value Dth, this corresponds to the case where there is no misrecognition in either the second left-side road-dividing line information or the second right-side road-dividing line information. Therefore, the road-dividing line information determination unit 103 outputs both the second left-side road-dividing line information and the second right-side road-dividing line information as the best road-dividing line information and proceeds to the processing of step S111.

[0061] If it is determined in step S110 that the deviation of the right-side road dividing line information (right-side deviation RDv) is greater than or equal to the threshold value Dth, then in step S109, the vehicle position estimation unit 104 estimates the vehicle position using only the second left-side road dividing line information, which is the best road dividing line information, and proceeds to processing in step S112.

[0062] If it is determined in step S110 that the deviation of the right-side road dividing line information (right-side deviation RDv) is less than the threshold value Dth, in step S111 the vehicle position estimation unit 104 estimates the vehicle position using both the second left-side road dividing line information and the second right-side road dividing line information, which are the best road dividing line information, and proceeds to processing in step S112.

[0063] Finally, in step S112, the driving route generation device 200 and the vehicle control device 300 perform vehicle control corresponding to the best road lane marking information, which is the vehicle position estimation result output from the vehicle position estimation unit 104 in any one of steps S108, S109, and S111. The above is an overview of the vehicle position estimation method using the vehicle position estimation device 100 according to the first embodiment.

[0064] If an error occurs in the first road-dividing line information, the first road-dividing line information will deviate significantly from the second road-dividing line information. Therefore, by performing the processing as described above, it is possible to select road-dividing line information so as not to use, for estimating the vehicle position, road-dividing line information that results in a large deviation between the first road-dividing line information acquired from the camera 90 and the second road-dividing line information based on the vehicle position information and map information acquired from the locator 91. As a result, it is possible to continue estimating the vehicle position with high accuracy even when the recognition situation for the road-dividing line information changes dynamically.

[0065] In other words, the vehicle position estimation device according to embodiment 1 dynamically changes the road-dividing line information used for vehicle position estimation depending on the occurrence state of erroneous recognition, thereby enabling continuous highly accurate vehicle position estimation as long as at least one piece of road-dividing line information can be correctly recognized.

[0066] By using the vehicle position estimation device 100 according to the first embodiment, even if the white lines that serve as landmarks for guiding the vehicle to the branching or merging lane are omitted or are faded, particularly at locations where branching and merging occur in the travel lane, it is more likely that the vehicle position estimation can be continued in the branching or merging lane.

[0067] <Advantages of First Embodiment> According to the vehicle position estimation device and vehicle position estimation method of embodiment 1, the degree of deviation between the first left-side road dividing line information and the second left-side road dividing line information, and the degree of deviation between the first right-side road dividing line information and the second right-side road dividing line information are each calculated, and the second road dividing line information is used to estimate the vehicle position only when the degree of deviation is less than a threshold value, thereby achieving the effect of obtaining a vehicle position estimation device and vehicle position estimation method that are capable of continuously estimating the vehicle position with high accuracy.

[0068] Embodiment 2 <Configuration of Vehicle Position Estimation Device According to Second Embodiment> 11 is a block diagram showing the configuration of a vehicle position estimation device 150 according to the second embodiment. The vehicle position estimation device 150 according to the second embodiment further includes a correction amount calculation unit 110 in addition to the configuration of the vehicle position estimation device 100 according to the first embodiment. The vehicle position estimation device 150 outputs the vehicle position estimation result to the travel route generation device 200, and the travel route generation device 200 outputs the generated travel route to the vehicle control device 300. Only the correction amount calculation unit 110, which is different from the configuration of the vehicle position estimation device 100 according to the first embodiment, will be described below.

[0069] The correction amount calculation unit 110 calculates three types of correction amounts in the following vehicle position estimation process, for example, for first road dividing line information based on an image of the area around the vehicle 10 captured by the camera 90, and second road dividing line information based on the vehicle position information and map information obtained from the locator 91, and outputs the results to the road dividing line information determination unit 103. (1) First correction amount for vehicle position estimation using road marking information on both the left and right sides (2) Second correction amount for vehicle position estimation using only left lane marking information (3) Third correction amount for vehicle position estimation using only right-side lane marking information Regarding the above three types of correction amounts, case (1) will be explained using FIG. 12, case (2) will be explained using FIG. 13, and case (3) will be explained using FIG.

[0070] <(1) First correction amount for vehicle position estimation using road marking information on both the left and right sides> FIG. 12 is a schematic diagram illustrating a first correction amount for estimating the vehicle position using road dividing line information on both the left and right sides.

[0071] As shown in the left and center diagrams in FIG. 12, for the first left-side point cloud data 25a (first left-side road-dividing line information) and the second left-side point cloud data 26a (second left-side road-dividing line information), a process is performed to calculate the deviation Δd between points in the point cloud data that are the same distance from each other in the longitudinal direction of the vehicle 10 for all point cloud data included in the distance range represented by both the first left-side road-dividing line information and the second left-side road-dividing line information. Specifically, the deviation Δd is calculated sequentially for each point within the same distance range. Then, the sum Δdsum of the absolute values ​​of the deviation Δd, i.e., the left deviation degree LDv of the left-side road-dividing line information, is obtained.

[0072] The same processing as for the left-side road-dividing line information is performed on the first right-side point cloud data 25b, which is the first right-side road-dividing line information, and the second right-side point cloud data 26b, which is the second right-side road-dividing line information. As a result, the right deviation degree RDv of the right-side road-dividing line information is obtained.

[0073] A first correction amount is calculated using both the left deviation degree LDv and the right deviation degree RDv. Specifically, the correction amount is expressed as a rotation angle. The rotation angle, which is the first correction amount, is applied to convert second left-side point cloud data 26a and second right-side point cloud data 26b, which are the second road-dividing line information, to obtain second left-side corrected point cloud data 27a and second right-side corrected point cloud data 27b after the first correction amount has been applied. The diagram on the right side of FIG. 12 shows, as an example, second left-side corrected point cloud data 27a and second right-side corrected point cloud data 27b when the first correction amount is a rotation angle of 7°.

[0074] <(2) Second correction amount for vehicle position estimation using only left-side lane marking information> FIG. 13 is a schematic diagram illustrating a second correction amount for estimating the vehicle position using only the left-side lane-dividing line information.

[0075] As shown in the left and center diagrams in FIG. 13, for the first left-side point cloud data 25a (first left-side road-dividing line information) and the second left-side point cloud data 26a (second left-side road-dividing line information), a process is performed to calculate the deviation Δd between points in the point cloud data that are the same distance from each other in the longitudinal direction of the vehicle 10 for all point cloud data included in the distance range represented by both the first left-side road-dividing line information and the second left-side road-dividing line information. Specifically, the deviation Δd is calculated sequentially for each point within the same distance range. Then, the sum Δdsum of the absolute values ​​of the deviation Δd, i.e., the left deviation degree LDv of the left-side road-dividing line information, is obtained.

[0076] The left deviation LDv is used to calculate a second correction amount. The second left-side point cloud data 26a, which is the second road-dividing line information, is converted using the rotation angle as the second correction amount to generate second left-side corrected point cloud data 27a after the second correction amount has been applied. The diagram on the right side of FIG. 13 shows, as an example, second left-side corrected point cloud data 27a when the second correction amount is set to a rotation angle of -15°.

[0077] <(3) Third correction amount for vehicle position estimation using only right-side lane marking information> FIG. 14 is a schematic diagram illustrating the third correction amount for estimating the vehicle position using only the right-side lane-dividing line information.

[0078] As shown in the left and center diagrams in FIG. 14, for the first right-side point cloud data 25b (first right-side road-dividing line information) and the second right-side point cloud data 26b (second right-side road-dividing line information), a process is performed to calculate the deviation Δd between points in the point cloud data that are the same distance from each other in the longitudinal direction of the vehicle 10 for all point cloud data included in the distance range represented by both the first right-side road-dividing line information and the second right-side road-dividing line information. Specifically, the deviation Δd is calculated sequentially for each point within the same distance range. Then, the sum Δdsum of the absolute values ​​of the deviation Δd, i.e., the right deviation degree RDv of the right-side road-dividing line information, is obtained.

[0079] The right deviation degree RDv is used to calculate a third correction amount. The rotation angle, which is the third correction amount, is applied to convert the second right-side point cloud data 26b, which is the second road-dividing line information, to obtain second right-side corrected point cloud data 27b after the third correction amount is applied. Figure 14 The diagram on the right side of the middle shows, as an example, second right-side corrected point cloud data 27b when the third correction amount is set to a rotation angle of 0.5°.

[0080] By performing the above processes, three types of correction amounts are obtained: a first correction amount using both left-side road-dividing line information and right-side road-dividing line information; a second correction amount using only left-side road-dividing line information; and a third correction amount using only right-side road-dividing line information. In the above example, the third correction amount, which has the smallest rotation angle among the correction amounts, can be evaluated as having the smallest deviation. In other words, the second right-side road-dividing line information is output as the best road-dividing line information. The above series of steps is shown in Figure 15. In short, the correction amount calculation unit 110 calculates at least one of the correction angle in the rotation process and the translation amount in the translation process as the first correction amount, the second correction amount, and the third correction amount.

[0081] <Method for estimating vehicle position according to second embodiment> A vehicle position estimation method using the vehicle position estimation device 150 according to the second embodiment will be described below with reference to the flowchart in Fig. 16. Note that the final step, step S210, corresponds to the operations of the driving route generation device 200 and the vehicle control device 300 based on the output of the vehicle position estimation device 150.

[0082] First, in step S201, the first road dividing line information acquisition unit 101 photographs the surroundings of the vehicle 10 using the camera 90, outputs first road dividing line information consisting of first left side road dividing line information and first right side road dividing line information, which are position information of the road dividing lines on the left and right sides of the lane 20 in which the vehicle 10 is traveling, and proceeds to processing in step S202.

[0083] In step S202, the second road dividing line information acquisition unit 102 outputs second road dividing line information consisting of second left side road dividing line information and second right side road dividing line information, which are position information of the road dividing lines on the left and right sides of the lane 20 in which the vehicle 10 is traveling, based on the vehicle position information and map information acquired by the locator 91, and proceeds to processing in step S203.

[0084] In step S203, the correction amount calculation unit 110 calculates a first correction amount when the vehicle position estimation process is performed using the road dividing line information on both the left and right sides, and the process proceeds to step S204.

[0085] In step S204, the correction amount calculation unit 110 calculates a second correction amount when the vehicle position estimation process is performed using only the left-side lane dividing line information, and the process proceeds to step S205.

[0086] In step S205, the correction amount calculation unit 110 calculates a third correction amount when the vehicle position estimation process is performed using only the right-side lane dividing line information, and the process proceeds to step S206.

[0087] In step S206, the lane-dividing line information determination unit 103 determines whether the minimum correction amount, which is the smallest value among the three types of correction amounts calculated by the correction amount calculation unit 110, is equal to or greater than a threshold value (threshold correction amount DAth). If the minimum correction amount is equal to or greater than the threshold correction amount DAth, the process proceeds to step S211, and if the minimum correction amount is less than the threshold correction amount DAth, the process proceeds to step S207.

[0088] In step S207, the lane-dividing line information determination unit 103 determines whether the first correction amount using the lane-dividing line information on both the left and right sides is the smallest of the three types of correction amounts calculated by the correction amount calculation unit 110. If the first correction amount is the smallest, the process proceeds to step S213; if the first correction amount is not the smallest, the process proceeds to step S208.

[0089] In step S208, the lane-dividing line information determination unit 103 determines whether the third correction amount using the right-side lane-dividing line information is the smallest of the three types of correction amounts calculated by the correction amount calculation unit 110. If the third correction amount is the smallest, the process proceeds to step S212; if the third correction amount is not the smallest, the process proceeds to step S209.

[0090] In step S211, based on the result that "invalid" is output as the best road-lagging line information from the road-lagging line information determination unit 103, the vehicle position estimation unit 104 determines that the vehicle position cannot be estimated and does not output the vehicle position estimation result. This is because in step S206, if the minimum correction amount, which is the smallest value of the three correction amounts calculated by the correction amount calculation unit 110, is equal to or greater than the threshold correction amount DAth, the road-lagging line information determination unit 103 outputs "invalid" as the best road-lagging line information.

[0091] If it is determined in step S207 that the first correction amount using the road dividing line information on both the left and right sides is the smallest, in step S213 the vehicle position estimation unit 104 estimates the vehicle position using both the first left road dividing line information and the second right road dividing line information, which are the best road dividing line information, and proceeds to processing in step S210.

[0092] If it is determined in step S208 that the third correction amount using the right-side road dividing line information is not the smallest, in step S209, the vehicle position estimation unit 104 estimates the vehicle position using only the second left-side road dividing line information, which is the best road dividing line information, and proceeds to processing in step S210.

[0093] If it is determined in step S208 that the third correction amount using the right-side road dividing line information is the smallest, in step S212 the vehicle position estimation unit 104 estimates the vehicle position using only the second right-side road dividing line information, which is the best road dividing line information, and proceeds to processing in step S210.

[0094] Finally, in step S210, the driving route generation device 200 and the vehicle control device 300 perform vehicle control corresponding to the best road lane marking information, which is the vehicle position estimation result output from the vehicle position estimation unit 104 in any one of steps S209, S212, and S213. The above is an overview of the vehicle position estimation method using vehicle position estimation device 150 according to the second embodiment.

[0095] It is also possible to provide the vehicle position estimation unit 104 with the function of calculating and outputting the above-mentioned correction amount without providing the correction amount calculation unit 110, and have the vehicle position estimation unit 104 calculate the correction amount.

[0096] If an error occurs in the first road-dividing line information, the first road-dividing line information will deviate significantly from the second road-dividing line information based on the map information, i.e., the degree of deviation will increase. Therefore, if road-dividing line information with an error is used, the amount of correction will increase. Therefore, by performing the processing described above, road-dividing line information with an increased amount of correction, i.e., road-dividing line information with an error, can be selected so that the road-dividing line information with an error is not used for estimating the vehicle position. This makes it possible to continue high-precision vehicle position estimation processing even when the recognition situation of road-dividing line information changes dynamically.

[0097] <Advantages of the Second Embodiment> According to the vehicle position estimation device and vehicle position estimation method of embodiment 2, three types of correction amounts are calculated, and the combination of road dividing line information that results in the smallest correction amount is selected and used for vehicle position estimation, thereby achieving the effect of obtaining a vehicle position estimation device and vehicle position estimation method that are capable of continuously estimating the vehicle position with high accuracy.

[0098] Embodiment 3 In the vehicle position estimation device and vehicle position estimation method according to the third embodiment, the first road-dividing line information acquisition unit 101 acquires first road-dividing line information reliability, which is an index of the reliability of the first road-dividing line information, based on an image of the lane captured by the camera 90. The second road-dividing line information acquisition unit 102 acquires second road-dividing line information reliability, which is an index of the reliability of the second road-dividing line information, based on the vehicle position information and map information acquired by the locator 91.

[0099] The first road-dividing line information reliability refers to the reliability of the first left-hand road-dividing line information, which is the position information of the first left-hand road-dividing line, and the first right-hand road-dividing line information, which is the position information of the first right-hand road-dividing line. Similarly, the second road-dividing line information reliability refers to the reliability of the second left-hand road-dividing line information, which is the position information of the second left-hand road-dividing line, and the second right-hand road-dividing line information, which is the position information of the second right-hand road-dividing line.

[0100] The road-dividing line information determination unit 103 uses the first road-dividing line information reliability and the second road-dividing line information reliability to select a combination of road-dividing line information that will yield the best vehicle position estimation result from among the vehicle position estimation using the left road-dividing line, the vehicle position estimation using the right road-dividing line, and the vehicle position estimation using the left road-dividing line and the right road-dividing line, based on the first left-side road-dividing line information and the first left-side road-dividing line information reliability and the first right-side road-dividing line information reliability and the second left-side road-dividing line information and the second right-side road-dividing line information reliability. The road-dividing line information obtained as a result of the selection is output as best road-dividing line information.

[0101] For example, if the white line that is the first left-side road-dividing line is faded, the reliability of the first left-side road-dividing line information is low, and therefore, when estimating the vehicle position, the vehicle position estimation using the left-side road-dividing line is not optimal. In other words, it is determined that the vehicle position estimation is not the best road-dividing line information.

[0102] In addition, if the reliability of the first road-dividing line information and the reliability of the second road-dividing line information are both low and cannot be used to estimate the vehicle's position, the road-dividing line information determination unit 103 determines that it is not possible to estimate the vehicle's position and outputs "invalid" as the best road-dividing line information.

[0103] If the best road-dividing line information output by the road-dividing line information determination unit 103 is valid, the vehicle position estimation unit 104 estimates the vehicle position by appropriately correcting the position of the second road-dividing line information using road-dividing line information on at least one of the left and right sides from the first road-dividing line information and the second road-dividing line information in accordance with the best road-dividing line information.

[0104] <Advantages of the Third Embodiment> According to the vehicle position estimation device and vehicle position estimation method of embodiment 3, the best road-dividing line information is obtained based on the first road-dividing line information reliability and the second road-dividing line information reliability, thereby achieving the effect of obtaining a vehicle position estimation device and vehicle position estimation method that are capable of continuously estimating the vehicle position with high accuracy.

[0105] Embodiment 4 <Configuration of Vehicle Control System According to Fourth Embodiment> 17 is a block diagram showing the configuration of a vehicle control system 500 according to Embodiment 4. The vehicle control system 500 includes the host vehicle position estimation device 100 according to Embodiment 1, a driving route generation device 200, and a vehicle control device 300.

[0106] The vehicle position estimation device 100 outputs the vehicle position estimation result of the vehicle 10 to the driving route generation device 200. Note that the vehicle position estimation device 100 according to the first embodiment may be replaced by the vehicle position estimation device 150 according to the second embodiment.

[0107] The driving route generation device 200 generates a driving route for the vehicle 10 to reach the destination point, using the vehicle position estimation result of the vehicle 10 output from the vehicle position estimation device 100. Note that a known method can be applied to generate the driving route.

[0108] The vehicle control device 300 sets a target trajectory and a target vehicle speed, which are target control amounts required for the vehicle 10 to travel on the travel route generated by the travel route generation device 200, and further calculates a target steering amount and a target acceleration / deceleration rate required to follow the target trajectory and the target vehicle speed. Note that known calculation methods can be applied to calculate the target steering amount and the target acceleration / deceleration rate. This concludes the description of the configuration of vehicle control system 500.

[0109] Hereinafter, the vehicle control of the vehicle 10 by the vehicle control system 500 will be described. The target control amounts calculated in the vehicle control device 300 of the vehicle control system 500, that is, the target steering amount and the target acceleration / deceleration, are output to the actuator 530, and automatic driving control of the vehicle 10 is executed.

[0110] The actuator 530 includes an Electronic Power Steering (EPS) controller 531 , a powertrain controller 532 , a brake controller 533 , an EPS unit 535 , a powertrain unit 536 , and a brake unit 537 .

[0111] The actuator 530 controls the EPS, brake, and accelerator so that the vehicle 10 follows the target steering amount and target acceleration / deceleration.

[0112] The EPS controller 531 controls the EPS unit 535 based on the target steering amount output from the vehicle control system 500. The EPS controller 531 can control, for example, the steering angle so that the vehicle 10 travels along a target trajectory.

[0113] Powertrain controller 532 controls powertrain unit 536 to achieve the target acceleration / deceleration output from vehicle control system 500. When the driver controls the speed instead of the automatic driving control, powertrain unit 536 is controlled based on the accelerator pedal depression amount.

[0114] The brake controller 533 controls the brake unit 537 so as to achieve the target acceleration / deceleration output from the vehicle control system 500. When the driver controls the speed instead of the automatic driving control, the brake controller 533 controls the brake unit 537 based on the amount of depression of the brake pedal.

[0115] <Advantages of the Fourth Embodiment> As described above, according to the vehicle control system of embodiment 4, the vehicle position information is calculated with high accuracy by the vehicle position estimation device of embodiment 1 or 2, thereby achieving the effect of realizing highly stable vehicle control based on the highly accurate vehicle position information.

[0116] The above has described a configuration in which the functions of the components of the vehicle position estimation devices 100 and 150 and the vehicle control system 500 according to the first to fourth embodiments are realized either by hardware or software, etc. However, the present invention is not limited to this, and some of the components of the vehicle position estimation devices 100 and 150 and the vehicle control system 500 may be realized by dedicated hardware, and other components may be realized by software, etc.

[0117] For example, as shown in FIGS. 18 and 19, some components are realized by a processing circuit 800 as dedicated hardware, and other components are realized by a processing circuit 800 as a processor 801. storage device The functions can be realized by reading and executing the program stored in 802 for causing a computer or the like to execute the vehicle position estimation method according to the first to third embodiments.

[0118] Furthermore, as shown in FIG. 19, the setting data used by each functional unit of the vehicle position estimation devices 100 and 150 is stored in a recording medium 803 that stores a part of the software, that is, a program 804 for causing a computer or the like to execute the vehicle position estimation method according to the first to third embodiments. storage device It may be installed on 802.

[0119] As described above, the vehicle position estimation devices 100 and 150 and the vehicle control system 500 according to the first to fourth embodiments can realize the above-described functions by hardware, software, or a combination of these.

[0120] <Summary of various aspects of the present application> Various aspects of the present application will be summarized below as appendices.

[0121] (Appendix 1) A vehicle position estimation device that estimates the position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, a first road-dividing line information acquisition unit that acquires first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; a second road-dividing line information acquisition unit that acquires second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; a road-dividing line information determination unit that determines a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; a vehicle position estimation unit that estimates the vehicle position by correcting the second road-dividing line information using a combination of the road-dividing line information determined by the road-dividing line information determination unit; and A vehicle position estimation device comprising:

[0122] (Appendix 2) The vehicle position estimation device described in Appendix 1, characterized in that the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position based on at least one of the degree of deviation between the first left-side road-dividing line information and the second left-side road-dividing line information regarding left-side road-dividing line information and the degree of deviation between the first right-side road-dividing line information and the second right-side road-dividing line information regarding right-side road-dividing line information.

[0123] (Appendix 3) the first road-dividing line information acquisition unit acquires, as the first road-dividing line information, first point cloud data consisting of first left-side point cloud data and first right-side point cloud data; the second road-dividing line information acquisition unit acquires, as the second road-dividing line information, second point cloud data consisting of second left-side point cloud data and second right-side point cloud data; The vehicle position estimation device according to claim 2, wherein the deviation is calculated using at least one of an inter-point distance between corresponding points in the first left-side point cloud data and the second left-side point cloud data, and an inter-point distance between corresponding points in the first right-side point cloud data and the second right-side point cloud data.

[0124] (Appendix 4) the first road-dividing line information acquisition unit acquires, as the first road-dividing line information, first point cloud data consisting of first left-side point cloud data and first right-side point cloud data; the second road-dividing line information acquisition unit acquires, as the second road-dividing line information, second point cloud data consisting of second left-side point cloud data and second right-side point cloud data; 3. The vehicle position estimation device according to claim 2, wherein the deviation is calculated using at least one of the inter-point distances between corresponding points among the first left-side point cloud data and the second left-side point cloud data when the centroid positions of the points of the first left-side point cloud data and the second left-side point cloud data are matched, and the inter-point distances between corresponding points among the first right-side point cloud data and the second right-side point cloud data when the centroid positions of the points of the first right-side point cloud data and the second right-side point cloud data are matched.

[0125] (Appendix 5) a correction amount calculation unit that calculates at least two of a first correction amount calculated when a combination of the first left-side road-dividing line information, the first right-side road-dividing line information, the second left-side road-dividing line information, and the second right-side road-dividing line information is used; a second correction amount calculated when a combination of the first left-side road-dividing line information and the second left-side road-dividing line information is used; and a third correction amount calculated when a combination of the first right-side road-dividing line information and the second right-side road-dividing line information is used, The vehicle position estimation device described in Appendix 1, characterized in that the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position based on two or more correction amounts calculated by the correction amount calculation unit.

[0126] (Appendix 6) the correction amount calculation unit calculates, as the first correction amount, the second correction amount, and the third correction amount, at least one of a correction angle during a rotation process and a translation amount during a translation process; The vehicle position estimation device described in Appendix 5, characterized in that the vehicle position estimation unit estimates the vehicle position by correcting the second point cloud data included in the second road-dividing line information acquired by the second road-dividing line information acquisition unit by performing the rotation processing and the translation processing.

[0127] (Appendix 7) the first road-delimiting line information acquisition unit further acquires first road-delimiting line information reliability, which is an index of reliability of the first road-delimiting line information; the second road-delimiting line information acquisition unit further acquires second road-delimiting line information reliability, which is an index of reliability of the second road-delimiting line information; The vehicle position estimation device described in Appendix 1 is characterized in that the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position using at least one of the first road-dividing line information reliability and the second road-dividing line information reliability.

[0128] (Appendix 8) 8. The vehicle position estimation device according to claim 1, wherein the second road dividing line information is road dividing line information based on map information.

[0129] (Appendix 9) A vehicle position estimation device according to any one of appendixes 1 to 8, which estimates a vehicle position based on road lane marking information; a driving route generation device that generates a driving route for the vehicle to reach a target point based on the vehicle position output from the vehicle position estimation device; a vehicle control device that sets a target trajectory and a target vehicle speed to be used when performing vehicle control of the vehicle on the generated travel route; A vehicle control system comprising:

[0130] (Appendix 10) A vehicle position estimation method for estimating a vehicle position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, using a vehicle position estimation device, comprising: acquiring first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; acquiring second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; determining a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; estimating the vehicle position by correcting the second road-dividing line information using the determined combination of road-dividing line information; A vehicle position estimation method comprising:

[0131] (Appendix 11) The vehicle position estimation method according to claim 10, wherein the second road dividing line information is road dividing line information based on map information.

[0132] Although the present disclosure describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to application to a particular embodiment, but may be applied to the embodiments alone or in various combinations.

[0133] Therefore, countless variations not illustrated are conceivable within the scope of the technology disclosed in the present specification, including, for example, cases where at least one component is modified, added, or omitted, and cases where at least one component is extracted and combined with components of another embodiment. [Explanation of symbols]

[0134] 10 vehicle, 20 lane, 21 first left road dividing line, 22 first right road dividing line, 23 second left road dividing line, 24 second right road dividing line, 25 first point cloud data, 25a first left point cloud data, 25b first right point cloud data, 26 second point cloud data, 26a second left point cloud data, 26b second right point cloud data, 27a second left corrected point cloud data, 27b second right corrected point cloud data, 30a, 31a center of gravity position, 90 camera, 91 locator, 100, 150 vehicle position estimation device, 101 first road dividing line information acquisition unit, 102 second road dividing line information acquisition unit, 103 road dividing line information determination unit, 104 vehicle position estimation unit, 110 correction amount calculation unit, 200 driving path generation device, 300 vehicle control device, 500 vehicle control system, 530 actuator 、5 31 EPS controller, 532 powertrain controller, 533 brake controller, 535 EPS unit, 536 powertrain unit, 537 brake unit, 800 processing circuit, 801 processor, 802 storage device , 803 recording medium, 804 program

Claims

1. A vehicle position estimation device that estimates the position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, a first road-dividing line information acquisition unit that acquires first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; a second road-dividing line information acquisition unit that acquires second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; a road-dividing line information determination unit that determines a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; a vehicle position estimation unit that estimates the vehicle position by correcting the second road-dividing line information using a combination of the road-dividing line information determined by the road-dividing line information determination unit; A vehicle position estimation device comprising:

2. 2. The vehicle position estimation device according to claim 1, wherein the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position based on at least one of the degree of deviation between the first left-side road-dividing line information and the second left-side road-dividing line information regarding left-side road-dividing line information and the degree of deviation between the first right-side road-dividing line information and the second right-side road-dividing line information regarding right-side road-dividing line information.

3. the first road-dividing line information acquisition unit acquires, as the first road-dividing line information, first point cloud data consisting of first left-side point cloud data and first right-side point cloud data; the second road-dividing line information acquisition unit acquires, as the second road-dividing line information, second point cloud data consisting of second left-side point cloud data and second right-side point cloud data; 3. The vehicle position estimation device according to claim 2, wherein the deviation is calculated using at least one of the inter-point distances between corresponding points in the first left-side point cloud data and the second left-side point cloud data, and the inter-point distances between corresponding points in the first right-side point cloud data and the second right-side point cloud data.

4. the first road-dividing line information acquisition unit acquires, as the first road-dividing line information, first point cloud data consisting of first left-side point cloud data and first right-side point cloud data; the second road-dividing line information acquisition unit acquires, as the second road-dividing line information, second point cloud data consisting of second left-side point cloud data and second right-side point cloud data; 3. The vehicle position estimation device according to claim 2, wherein the deviation is calculated using at least one of the inter-point distances between corresponding points among the first left-side point cloud data and the second left-side point cloud data when the center of gravity positions of the points of the first left-side point cloud data and the second left-side point cloud data are matched, and the inter-point distances between corresponding points among the first right-side point cloud data and the second right-side point cloud data when the center of gravity positions of the points of the first right-side point cloud data and the second right-side point cloud data are matched.

5. a correction amount calculation unit that calculates at least two of a first correction amount calculated when a combination of the first left-side road-dividing line information, the first right-side road-dividing line information, the second left-side road-dividing line information, and the second right-side road-dividing line information is used; a second correction amount calculated when a combination of the first left-side road-dividing line information and the second left-side road-dividing line information is used; and a third correction amount calculated when a combination of the first right-side road-dividing line information and the second right-side road-dividing line information is used, 2. The vehicle position estimation device according to claim 1, wherein the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position based on two or more correction amounts calculated by the correction amount calculation unit.

6. the correction amount calculation unit calculates, as the first correction amount, the second correction amount, and the third correction amount, at least one of a correction angle during a rotation process and a translation amount during a translation process; The vehicle position estimation device according to claim 5, characterized in that the vehicle position estimation unit estimates the vehicle position by correcting the second point cloud data included in the second road dividing line information acquired by the second road dividing line information acquisition unit by performing the rotation processing and the parallel translation processing.

7. the first road-dividing line information acquisition unit further acquires first road-dividing line information reliability, which is an index of reliability of the first road-dividing line information; the second road-dividing line information acquisition unit further acquires second road-dividing line information reliability, which is an index of reliability of the second road-dividing line information; 2. The vehicle position estimation device according to claim 1, wherein the road-dividing line information determination unit determines a combination of road-dividing line information to be used for estimating the vehicle position, using at least one of the first road-dividing line information reliability and the second road-dividing line information reliability.

8. 8. The vehicle position estimation device according to claim 1, wherein the second road dividing line information is road dividing line information based on map information.

9. a vehicle position estimation device according to any one of claims 1 to 7, which estimates a vehicle position based on road division line information; a driving route generation device that generates a driving route for the vehicle to reach a target point based on the vehicle position output from the vehicle position estimation device; a vehicle control device that sets a target trajectory and a target vehicle speed to be used when performing vehicle control of the vehicle on the generated travel route; A vehicle control system comprising:

10. A vehicle position estimation method for estimating a vehicle position of a vehicle traveling on a road defined by a left-side road dividing line and a right-side road dividing line, using a vehicle position estimation device, comprising: acquiring first road-dividing line information including first left-side road-dividing line information representing the left-side road-dividing line and first right-side road-dividing line information representing the right-side road-dividing line detected by an imaging element mounted on the vehicle; acquiring second road-dividing line information including second left-side road-dividing line information representing the left-side road-dividing line and second right-side road-dividing line information representing the right-side road-dividing line, both of which are acquired by a method different from the detection method using the image sensor; determining a combination of road-dividing line information to be used for estimating the vehicle position based on the first road-dividing line information and the second road-dividing line information; estimating the vehicle position by correcting the second road-dividing line information using the determined combination of road-dividing line information; A vehicle position estimation method comprising:

11. 11. The method for estimating the vehicle position according to claim 10, wherein the second lane dividing line information is lane dividing line information based on map information.

Images