Travel path estimation method and travel path estimation device

Abstract

To early determine an increase in the number of lanes to estimate an appropriate course even in a place where the number of lanes ahead of a cross-point is increased.SOLUTION: A course estimation method includes: acquiring a distance from user's vehicle to a cross-point being ahead of a road along which the vehicle is traveling; detecting compartment lines positioned in the left and the right of the vehicle; calculating a degree of parallelism between the detected left and right compartment lines; setting a parallelism threshold for determining whether the compartment lines positioned in the left and the right of the vehicle are parallel with each other or not to a reference value if the distance from the vehicle to the cross-point is equal to or longer than a prescribed determination distance; and setting the parallelism threshold to a value higher than the reference value if the distance from the vehicle to the cross-point is shorter than the determination distance; determining that the left and right compartment lines are parallel with each other if the calculated degree of parallelism is higher than the set parallelism threshold; and estimating a course along which the vehicle is to travel, on the basis of the left and right compartment lines if it is determined that the left and right compartment lines are parallel with each other.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a roadway estimation method and a roadway estimation device. [Background technology]

[0002] Conventionally, a lane marking recognition device that recognizes lane markings painted on the road on which a vehicle is traveling is disclosed in Patent Document 1. When the lane marking recognition device disclosed in Patent Document 1 determines that the lane marking candidates on the left and right sides of the vehicle are not parallel, it excludes the lane marking candidate on the left and right sides of the vehicle that has the greater difference with respect to the direction of the lane on which the vehicle is traveling. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-191741 Summary of the Invention [Problem to be solved by the invention]

[0004] In the conventional lane marking recognition device described above, in order to determine whether lane marks are parallel, it is necessary to set a threshold value taking into account detection error. In particular, because detection error increases as the lane marking of the vehicle becomes more distant, it is necessary to set a threshold value with a margin of error. However, in places where non-parallel lane markings exist due to an increase in the number of lanes, such as before an intersection, it is necessary to quickly determine whether the lane markings are non-parallel. However, if the threshold value is set with a margin of error, it is not possible to quickly determine whether the number of lanes has increased.

[0005] Therefore, the present invention has been proposed in consideration of the above-mentioned situation, and aims to provide a route estimation method and device that can quickly determine an increase in the number of lanes and estimate an appropriate route, even in places where the number of lanes increases just before an intersection. [Means for solving the problem]

[0006] To solve the above-mentioned problems, a lane estimation method and device according to one aspect of the present invention detects lane markings located on the left and right of a vehicle and calculates a parallelism indicating the degree to which the detected lane markings are parallel. If the distance from the vehicle to the intersection is equal to or greater than a predetermined determination distance, a parallelism threshold for determining whether the lane markings located on the left and right of the vehicle are parallel is set to a reference value. On the other hand, if the distance from the vehicle to the intersection is less than the determination distance, the parallelism threshold is set higher than the reference value. As a result, if the calculated parallelism is higher than the set parallelism threshold, the lane markings are determined to be parallel, and if the lane markings are determined to be parallel, the lane markings on which the vehicle will travel are estimated based on the lane markings. [Effects of the Invention]

[0007] According to the present invention, even in a location where the number of lanes increases before an intersection, it is possible to determine the increase in the number of lanes early and estimate an appropriate driving route. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing the configuration of a vehicle system equipped with a roadway estimation device according to the first embodiment. [Figure 2] FIG. 2 is a diagram for explaining a method for calculating the parallelism by the travel path estimation device according to the first embodiment. [Figure 3] FIG. 3 is a diagram for explaining a method for calculating the parallelism by the travel path estimation device according to the first embodiment. [Figure 4] FIG. 4 is a diagram for explaining a method for calculating the parallelism by the travel path estimation device according to the first embodiment. [Figure 5] FIG. 5 is a flowchart showing the processing procedure of the travel path estimation process performed by the travel path estimation device according to the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining a method for estimating a traveling path by the traveling path estimation device according to the first embodiment. [Figure 7]FIG. 7 is a diagram for explaining a method for determining the direction of the lane in which the vehicle is traveling by the travel path estimation device according to the first embodiment. [Figure 8] FIG. 8 is a diagram for explaining a method for determining the direction of the lane in which the vehicle is traveling by the roadway estimation device according to the first embodiment. [Figure 9] FIG. 9 is a diagram for explaining a method for determining the direction of the lane in which the vehicle is traveling by the roadway estimation device according to the first embodiment. [Figure 10A] FIG. 10A is a diagram for explaining a method for determining the direction of the lane in which the vehicle is traveling by the travel path estimation device according to the first embodiment. [Figure 10B] FIG. 10B is a diagram for explaining a method for determining the direction of the lane in which the vehicle is traveling by the roadway estimation device according to the first embodiment. [Figure 11] FIG. 11 is a diagram for explaining a method for setting the parallelism threshold value by the lane estimation device according to the first embodiment. [Figure 12] FIG. 12 is a diagram for explaining a method for setting the parallelism threshold value by the lane estimation device according to the first embodiment. [Figure 13] FIG. 13 is a flowchart showing the processing procedure of the travel path estimation process performed by the travel path estimation device according to the second embodiment. [Figure 14] FIG. 14 is a diagram for explaining a method for estimating a traveling path by the traveling path estimation device according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] [First embodiment] A first embodiment of the present invention will be described below with reference to the drawings. In the description of the drawings, the same parts are given the same reference numerals and detailed description will be omitted.

[0010] [Vehicle system configuration] Fig. 1 is a block diagram showing the configuration of a vehicle system of a vehicle equipped with a roadway estimation device according to this embodiment. As shown in Fig. 1, the vehicle system 100 includes a roadway estimation device 1, a map information receiving unit 3, a sensor 5, a GNSS receiving unit 7, and an actuator 9. The vehicle system 100 is installed in a vehicle capable of autonomous driving or a vehicle that provides driving assistance to the driver.

[0011] The lane estimation device 1 detects lane markings located on the left and right of the vehicle, determines whether the detected lane markings are parallel, and if they are determined to be parallel, estimates the lane on which the vehicle will travel based on the lane markings. At this time, the lane estimation device 1 calculates a parallelism indicating the degree to which the lane markings are parallel, and if the calculated parallelism is higher than a parallelism threshold, determines that the lane markings are parallel.

[0012] In particular, the lane estimation device 1 sets the parallelism threshold to a reference value when the distance from the vehicle to the intersection is equal to or greater than a predetermined judgment distance, and sets the parallelism threshold higher than the reference value when the distance from the vehicle to the intersection is less than the judgment distance. This allows for stricter judgment of whether the left and right lane markings are parallel as the vehicle approaches the intersection, making it possible to quickly determine the increase in the number of lanes and estimate an appropriate lane, even in places where the number of lanes increases before the intersection.

[0013] The map information receiving unit 3 acquires map information from an external server or the like. For example, the map information includes a three-dimensional high-precision map, a navigation map (navigation map) that is less accurate than a high-precision map and covers a wide area, a global map, etc. The acquired map information is output to the roadway estimation device 1 and recorded in a storage device such as a memory (not shown).

[0014] The high-precision map includes, as information for each lane, information on lane nodes that indicate reference points on lane reference lines (for example, the center line within a lane) and information on lane links that indicate the section configuration of the lane between the lane nodes. The navigation map is equivalent to map data that holds node information and link information for each road, which is less accurate than the high-precision map and contains less information indicating road shape than the high-precision map. However, both the high-precision map and the navigation map can be applied to the lane estimation device 1 of this embodiment.

[0015] The map information may be obtained from an external server, or may be created by the vehicle itself based on the sensing results of the sensor 5, such as radar or LiDAR. The map information includes information on the lane the vehicle will travel in and intersection information, such as the distance to the intersection. The lane information includes dividing lines indicating the boundaries of the lane the vehicle will travel in, road shape, and route information. The intersection information includes information such as the distance and position to the intersection, and whether a right or left turn is possible.

[0016] The sensor 5 includes, for example, a camera, a radar sensor, etc. The camera may be, for example, a CCD (Charge Coupled Devices) camera or a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The camera may also be a monocular camera such as a near-infrared camera, or a stereo camera. Only one camera or multiple cameras may be installed on the host vehicle. The camera is attached at a predetermined height in the center of the vehicle width direction, and continuously captures images of an area including the road ahead of the host vehicle at a predetermined frame rate. The camera sequentially outputs the captured images as sensing information to the lane estimation device 1, and the sensing information is recorded in a storage device such as a memory (not shown).

[0017] The radar sensor includes a radar and a LiDAR (Light Detection and Ranging). The radar and LiDAR are sensors mounted on the vehicle to detect objects present around the vehicle and measure the distance to each point on the detected object. Note that the types and numbers of cameras, radar sensors, etc. are not limited to those described above. The distance measured by the radar sensor is output to the path estimation device 1 and recorded in a storage device such as a memory (not shown).

[0018] The GNSS receiver 7 receives GNSS signals from GNSS (Global Navigation Satellite System) satellites and outputs them to the path estimation device 1. The GNSS signals are recorded in a storage device such as a memory (not shown).

[0019] The actuator 9 drives the steering wheel, accelerator, and brake device of the host vehicle in response to control signals from the path estimation device 1. As a result, the actuator 9 controls the steering direction and steering amount of the host vehicle, the accelerator opening of the host vehicle, and the braking operation of the host vehicle's brake device, thereby generating vehicle behavior of the host vehicle.

[0020] As shown in FIG. 1, the roadway estimation device 1 includes a vehicle position estimation unit 11, an intersection information acquisition unit 13, a lane marking detection unit 15, a parallelism calculation unit 17, a parallelism threshold setting unit 19, a parallel determination unit 21, a roadway estimation unit 23, and a vehicle control unit 25.

[0021] The vehicle position estimation unit 11 estimates the position of the vehicle based on the map information received from the map information receiving unit 3 and the sensor data acquired from the sensor 5. For example, the vehicle position can be estimated by performing a matching process on the target information included in the map information and the target information included in the sensor data.

[0022] The vehicle position estimation unit 11 may also estimate the vehicle's own position based on the GNSS signal from the GNSS receiver 7 and the sensor data from the sensor 5. Furthermore, the vehicle position estimation unit 11 may acquire the position on a navigation map where the vehicle is traveling. A Kalman filter using the GNSS signal or the like may be used as a method for estimating the vehicle's position.

[0023] The intersection information acquisition unit 13 acquires intersection information about intersections ahead of the road on which the vehicle is traveling and information about the distance from the vehicle to the intersection, based on the map information received from the map information reception unit 3. An intersection refers to an area including an intersection between two or more roads and the surrounding area of ​​the intersection. The intersection to which this embodiment is applied may be only an intersection where a right turn or a left turn is possible.

[0024] In the intersection information, an intersection is defined as a location where nodes or links of roads traveling in different directions intersect, and the distance from the vehicle to the intersection ahead is calculated. Alternatively, a node, a link, or a collection of these may be defined as a road segment, and if any of them has a tag indicating an intersection or the presence or absence of a traffic light recorded, it may be defined as an intersection. The distance to the intersection can be calculated by tracing from the node or link where the vehicle is located to the node or link that has intersection information, and then deriving the cumulative distance between each node and link. The intersection information may include the road structure of the intersection, traffic light information, information about oncoming vehicles planning to pass through the intersection, information about whether a right or left turn is possible at the intersection, and the locations of other vehicles present at the intersection. The intersection information may be detected based on the vehicle's current position information detected by the GNSS receiver 7 and the map information from the map information receiver 3.

[0025] The lane marking detection unit 15 detects lane markings located on the left and right sides of the road on which the host vehicle is traveling from within the detection distance of the sensor 5. Specifically, the lane marking detection unit 15 detects left and right lane markings drawn with paint or the like as left and right lane markings from an image of the road ahead of the host vehicle captured by the camera of the sensor 5. The lane marking detection unit 15 may also extract potential lane markings on the road within the detection distance as candidates for left and right lane markings by using known image recognition processing such as pattern matching or line extraction Hough transform processing.

[0026] The lane marking detection unit 15 connects the candidates for the left and right lane marks recognized above to the lines recognized as lane marks in the previous image to create lines that will become lane marks. Specifically, the lane marking detection unit 15 connects the candidate lane marking that is closest to the lane marking recognized in the previous image to the end of the lane marking in the previous image to create a lane marking that is continuous with the lane marking recognized previously.

[0027] The parallelism calculation unit 17 calculates the parallelism of the left and right lane lines detected by the lane line detection unit 15. The parallelism indicates the degree of parallelism of the lane lines located on the left and right of the vehicle. Specifically, the parallelism calculation unit 17 converts the image of the road including the lane lines into a top view from a viewpoint above the vehicle using bird's-eye view transformation, and calculates the parallelism of the lane lines based on the converted top view.

[0028] For example, as shown in Fig. 2, the parallelism calculation unit 17 calculates the width between the left and right lane markings and calculates the parallelism from the difference between the widths. Specifically, the parallelism value is calculated using the following formula (1) using a reference width w1 of a lane marking near the vehicle as a reference, a comparative width w2 of a lane marking a certain distance ahead of the vehicle, and a constant r.

number

[0029] The reference width w1 and comparison width w2 are determined by converting the image captured by the sensor 5 camera into a top-down view from a viewpoint above the vehicle using bird's-eye view transformation and obtaining the widthwise distance between the left and right lane markings. The reference width w1 is the width of the lane the vehicle is traveling in just before the intersection. Alternatively, it may be a value calculated statistically from past detection results. The detection position of the comparison width w2 is detected within a range of 2 m (longer than the minimum detection distance) to 20 m ahead of the vehicle, and within a range of 10 m before the intersection to the intersection's stop line. The widths w1 and w2 between the lane markings are normalized (i.e., r = 3) so that the parallelism is 1 when the difference between w2 and w1 is 0 m, and 0 when the difference between w2 and w1 is 3.0 m (assuming the lane width is 3 m). Therefore, when the width of the dividing line does not increase (there is no difference between w2 and w1), the parallelism is high, and when the width of the dividing line increases (when the difference between w2 and w1 is large), the parallelism is low.

[0030] Alternatively, as shown in Figure 3, the direction vectors V1 and V2 from the start point to the end point of the lane markings L1 and L2 may be calculated, and the parallelism may be calculated from the angle and dot product between the direction vectors V1 and V2. In this case, the distance from the start point to the end point may be divided into multiple points, the direction vector between each point may be calculated, and the parallelism may be calculated based on the average value of the angle (0 to 180) and dot product (-1 to 1) between the paired direction vectors of the left and right lane markings. If the dot product is a negative value, the sign is reversed. The smaller the angle and the larger the dot product value, the more similar the directions, and therefore the higher the parallelism.

[0031] Alternatively, as shown in Figure 4, the normal vectors Vn1 and Vn2 at each point on the left and right lane lines L1 and L2 can be calculated, and the parallelism can be calculated based on the average value of the angle (0 to 180) between the normal vectors at the corresponding two points on the left and right lane lines L1 and L2 and the dot product (-1 to 1). If the dot product is a negative value, the sign is reversed. The smaller the angle and the larger the dot product, the higher the parallelism is set.

[0032] The parallelism threshold setting unit 19 sets the parallelism threshold to a reference value when the distance from the host vehicle to the intersection is equal to or greater than a predetermined determination distance. The parallelism threshold is a threshold for determining whether the left and right lane lines of the host vehicle are parallel. If the distance from the host vehicle to the intersection is less than the determination distance, the parallelism threshold is set higher than the reference value. The determination distance is set according to the length of an additional lane (a right-turn lane or a left-turn lane) before the intersection, and is the distance from the point where the additional lane starts to the intersection. In other words, when the host vehicle passes the point where the right-turn lane of the intersection starts, the parallelism threshold setting unit 19 sets the parallelism threshold higher than the reference value to more strictly determine whether the left and right lane lines are parallel. However, the determination distance may be the distance from the start point of the additional lane to the intersection plus the detection distance of the sensor 5.

[0033] For example, in the example of FIG. 2, the parallelism is normalized so that if the width of the lane markings does not increase (the difference between w2 and w1 is 0 [m]), the parallelism is 1, and if the width of the lane markings increases by one lane (assuming the lane width is 3 [m] and the difference between w2 and w1 is 3.0 [m]), the parallelism is 0. Therefore, the parallelism threshold is set to a value between 0 and 1. Therefore, the parallelism threshold setting unit 19 sets the reference value of the parallelism threshold to 0.7, and when the distance from the vehicle to the intersection is less than the judgment distance, the parallelism threshold is set to 0.9, which is higher than the reference value of 0.7. As a result, in a section before an intersection where the number of lanes is likely to increase, the parallelism threshold is increased to quickly determine whether the left and right lane markings are parallel, and the lane markings of the increasing lanes can be excluded, allowing the correct lane to be estimated. However, the parallelism threshold may also be set to gradually increase as the distance from the vehicle to the intersection decreases.

[0034] If the width of one lane marking is 0.2 m, there is a possibility of a difference of 0.4 m depending on whether the inside or outside of the lane marking is detected. Therefore, the minimum lane marking width (the difference between w2 and w1) can be set to 0.4 m, and the width when one lane is increased can be set to 3.0 m. This range can be normalized to calculate the parallelism. That is, r in equation (1) is normalized as 3.0 - 0.4 = 2.6. In this case, the reference value for the parallelism threshold is set to 0.77, which corresponds to when the difference between w2 and w1 is 1.0 m. If the difference is less than the judgment distance, the parallelism threshold is increased to 0.96, which corresponds to when the difference between w2 and w1 is 0.5 m.

[0035] The parallel determination unit 21 determines that the left and right lane markings are parallel if the parallelism calculated by the parallelism calculation unit 17 is higher than the parallelism threshold set by the parallelism threshold setting unit 19. On the other hand, if the calculated parallelism is equal to or lower than the set parallelism threshold, the parallel determination unit 21 determines that the left and right lane markings are not parallel. Images captured by the camera may contain objects that could be confused with lane markings, such as erased line marks, branch lines drawn along other lanes branching off from the vehicle's lane, road repair marks, or linear discolorations of parts of the road surface due to lighting conditions. Therefore, by performing a parallel determination, the system prevents objects that could be confused with lane markings from being mistakenly recognized as lane markings for the lane in which the vehicle is traveling.

[0036] When the parallel determination unit 21 determines that the left and right lane markings of the host vehicle are parallel, the lane estimation unit 23 estimates the lane on which the host vehicle will travel based on the left and right lane markings. In other words, when it is determined that the candidate lane markings on the left and right of the host vehicle are parallel, the lane estimation unit 23 estimates the candidate lane markings located on the left and right of the lane on which the host vehicle is traveling as the lane on which the host vehicle will travel.

[0037] On the other hand, if the lane estimation unit 23 determines that the left and right lane markings of the vehicle are not parallel, it sets the lane marking that is less different from the direction of the lane in which the vehicle is traveling as the reference lane marking, and estimates the lane on which the vehicle will travel based on the reference lane marking.

[0038] Specifically, the lane estimation unit 23 converts an image of a road including the lane markings on which the vehicle is traveling into a top view seen from a viewpoint above the vehicle by bird's-eye view transformation. Then, in the converted top view, the lane estimation unit 23 compares the inclination of the candidate lane markings on the left and right of the vehicle with the inclination of the direction in which the lane on which the vehicle is traveling extends, and selects the lane marking with the smaller difference in inclination and sets it as the reference lane marking.

[0039] At this time, the lane in which the vehicle is traveling may be estimated based on the past travel path of the vehicle, or the lane in which the vehicle is traveling may be estimated based on map information such as route information on a navigation map acquired from the map information receiving unit 3. Furthermore, the lane in which the vehicle is traveling may be estimated based on an arrow marking painted on the road surface in the lane in which the vehicle is traveling.

[0040] The vehicle control unit 25 provides various driving assistance to the driver of the vehicle based on information about the lane markings and road shape estimated as the vehicle's driving path by the driving path estimation unit 23. The driving assistance provided by the vehicle control unit 25 includes a lane keeping assist function. The lane keeping assist function automatically controls the steering device to make it easier for the vehicle to travel near the center of the lane when it is determined that there is a possibility that the vehicle will deviate from its lane.

[0041] The path estimation device 1 is a controller configured with general-purpose electronic circuits including a microcomputer, a microprocessor, and a CPU, and peripheral devices such as a memory, and is installed with a computer program for executing the path estimation process. Each function of the path estimation device 1 can be implemented by one or more processing circuits. The processing circuit may include, for example, a programmed processing device including an electrical circuit, or may include devices such as an application-specific integrated circuit (ASIC) or conventional circuit components arranged to perform the functions described in the embodiments.

[0042] [Route estimation processing] Next, a description will be given of the path estimation process executed by the path estimation device 1 according to this embodiment. Fig. 5 is a flowchart showing the processing procedure of the path estimation process executed by the path estimation device 1 according to this embodiment.

[0043] As shown in FIG. 5, in step S101, the vehicle position estimation unit 11 acquires the map information received by the map information receiving unit 3.

[0044] In step S103, the vehicle position estimation unit 11 acquires the current position of the vehicle received by the GNSS receiving unit 7 and the sensor data detected by the sensor 5, and compares them with the map information acquired in step S101 to estimate the vehicle's position.

[0045] In step S105, the intersection information acquisition unit 13 acquires intersection information about an intersection ahead of the road on which the host vehicle is traveling, based on the map information received from the map information receiving unit 3. In particular, the intersection information acquisition unit 13 acquires the distance from the host vehicle to the intersection ahead of the road on which the host vehicle is traveling. The intersection information acquisition unit 13 also acquires information about whether or not a right turn or a left turn is possible at the intersection ahead of the host vehicle.

[0046] In step S107, the lane marking detection unit 15 detects lane markings on the left and right of the vehicle, which are painted on the road by paint or the like, from an image of the road ahead of the vehicle captured by the camera of the sensor 5.

[0047] In step S109, the parallelism threshold setting unit 19 sets the parallelism threshold to the reference value when the distance from the host vehicle to the intersection is equal to or greater than a predetermined determination distance. That is, when the host vehicle is traveling before the point where the right-turn lane at the intersection starts, the parallelism threshold setting unit 19 sets the parallelism threshold to the reference value.

[0048] In step S111, the parallelism threshold setting unit 19 determines whether the distance from the host vehicle to the intersection is less than the determination distance. That is, the parallelism threshold setting unit 19 determines whether the host vehicle has passed the point where the right-turn lane of the intersection starts. If the distance from the host vehicle to the intersection is less than the determination distance, the process proceeds to step S113, and if the distance from the host vehicle to the intersection is not less than the determination distance, the process proceeds to step S115.

[0049] In step S113, the parallelism threshold setting unit 19 sets the parallelism threshold higher than the reference value because the distance from the host vehicle to the intersection is less than the determination distance. That is, the parallelism threshold setting unit 19 sets the parallelism threshold higher than the reference value because the host vehicle has passed the point where the right-turn lane at the intersection begins. This makes it possible to quickly determine that the left and right lane markings are not parallel, even if the right-turn lane begins before the intersection and the number of driving lanes increases, thereby preventing an erroneous estimation of a driving lane that bulges toward the added lane. Note that the parallelism threshold may be set higher than the reference value only if the intersection ahead of the host vehicle is an intersection where a right turn or a left turn is possible, based on the intersection information.

[0050] In step S115, the parallelism calculation unit 17 calculates the parallelism of the lane markings located on the left and right sides of the vehicle detected in step S107.

[0051] In step S117, the parallel determination unit 21 determines whether the parallelism calculated in step S115 is higher than the parallelism threshold set in step S109 or step S113. If the parallelism is higher than the parallelism threshold, it determines that the lane markings on the left and right of the vehicle are parallel, and the process proceeds to step S119. If the parallelism is equal to or lower than the parallelism threshold, it determines that the lane markings on the left and right of the vehicle are not parallel, and the process proceeds to step S121.

[0052] In step S119, the lane estimation unit 23 estimates the lane on which the host vehicle will travel based on the lane markings located on the left and right of the host vehicle. Specifically, if the lane markings located on the left and right of the host vehicle are parallel, the lane estimation unit 23 estimates the two lane markings located on the left and right as the lane on which the host vehicle will travel. In addition, the lane may be estimated to be the center line of the left and right lane markings, or one of the left and right lane markings may be moved to the center of the host vehicle and estimated to be the lane.

[0053] In step S121, the lane estimation unit 23 sets the lane markings located on the left and right of the vehicle that have the smaller difference from the direction of the lane in which the vehicle is traveling as the reference lane marking, and estimates the lane on which the vehicle will travel based on the reference lane marking.

[0054] For example, as shown in Figure 6, if it is determined that two lane markings L1 and L2 are not parallel, the lane marking L1 that is the least different from the direction of the lane the vehicle is traveling in is set as the reference lane marking. This reference lane marking is then moved by the lane width in the direction of the other lane marking L2 to set a virtual line VL1, and this virtual line VL1 and the reference lane marking L1 are estimated to be the road.

[0055] The direction of the lane in which the host vehicle is traveling may be set based on the travel path of the host vehicle. As shown in Figure 7, the direction indicated by the travel path R1 along which the host vehicle has traveled up to the current time is set as the direction of the lane in which the host vehicle is traveling. Then, by comparing the vectors V1 and V2 of the left and right lane markings L1 and L2 with the vector VR1 of the travel path R1, it is found that of the vectors V1 and V2, the vector V1 has the smaller difference from the vector VR1. Therefore, the lane marking L1 is set as the reference lane marking.

[0056] The direction of the lane in which the vehicle is traveling may also be set based on the direction of the road on which the vehicle is scheduled to travel. As shown in Fig. 8, the position of the intersection on which the vehicle is scheduled to travel is obtained from map information, and a vector VR2 indicating the direction from the current position to the intersection is calculated as the direction of the road on which the vehicle is scheduled to travel. Then, the direction of the road on which the vehicle is scheduled to travel may be set as the direction of the lane in which the vehicle is traveling.

[0057] Furthermore, the direction of the lane in which the host vehicle is traveling may be set based on the direction of the lane adjacent to the lane in which the host vehicle is traveling. As shown in FIG. 9, if there is an adjacent lane to the lane in which the host vehicle is traveling, the direction of the two dividing lines L1 and L3 of the adjacent lane may be set as the direction of the lane in which the host vehicle is traveling. As a method for determining whether or not there is an adjacent lane, it can be determined that there is an adjacent lane if the line type of the dividing line L1 is a dividing line that allows lane changes, such as a dashed line. It may also be determined that there is an adjacent lane if the width of the two dividing lines L1 and L3 of the adjacent lane is close to the lane width of the host vehicle and they are highly parallel to each other.

[0058] The direction of the lane in which the host vehicle is traveling may also be set based on the arrow markings of the lane in which the host vehicle is traveling. For example, as shown in Fig. 10A, if there is a straight-ahead and right-turn arrow marking 30 on the lane in which the host vehicle is traveling, the right-side lane marking L2 expands outward, so the direction of the left-side lane marking L1 is set as the direction of the lane in which the host vehicle is traveling. Similarly, as shown in Fig. 10B, if there is a straight-ahead and left-turn arrow marking 32 on the lane in which the host vehicle is traveling, the direction of the right-side lane marking L2 is set as the direction of the lane in which the host vehicle is traveling.

[0059] After the distance from the vehicle to the intersection becomes less than the determination distance, the following process may be further performed: The parallelism threshold setting unit 19 may set a switching distance that is shorter than the determination distance, and when the distance from the vehicle to the intersection becomes less than the switching distance, switch the parallelism threshold from a value higher than the reference value to the reference value.

[0060] For example, as shown in Fig. 11, the distance from point P2 where the lane addition is completed to the intersection is set as the switching distance K2. The switching distance K2 is shorter than the determination distance K1, which is set as the distance from point P1 where the right-turn lane starts to the intersection.

[0061] When the distance from the host vehicle to the intersection is less than the judgment distance K1, the left and right lane markings are likely not parallel due to the increase in the number of lanes, so the parallelism threshold needs to be set higher than the reference value. However, when the distance from the host vehicle to the intersection is less than the switching distance K2, the increase in the number of lanes has been completed, so the left and right lane markings are likely to be parallel, and there is no longer any need to set the parallelism threshold higher than the reference value. Therefore, the parallelism threshold setting unit 19 controls the parallelism threshold to be switched from a value higher than the reference value to the reference value when the distance from the host vehicle to the intersection is less than the switching distance K2.

[0062] In addition, the parallelism threshold setting unit 19 may switch the parallelism threshold from a value higher than the reference value to the reference value when a new lane marking is detected after the distance from the vehicle to the intersection becomes less than the judgment distance and it is determined that the lane markings on the left and right sides of the vehicle are not parallel.

[0063] As shown in Figure 12, after the distance from the host vehicle to the intersection becomes less than the judgment distance K1 and it is determined that the lane markings L1 and L2 on the left and right sides of the host vehicle are not parallel, a new lane marking L4 may be detected as the host vehicle continues to travel. In this case, the lane addition has been completed and the lane markings of the added lane have been detected, so it is no longer necessary to set the parallelism threshold higher than the reference value. Therefore, the parallelism threshold setting unit 19 controls the parallelism threshold to switch from a value higher than the reference value to the reference value.

[0064] Once the route along which the host vehicle will travel has been estimated in step S119 or step S121, the process proceeds to step S123. In step S123, the vehicle control unit 25 outputs a control signal to the actuator 9 so that the host vehicle travels along the route estimated in step S119 or step S121. Then, the actuator 9 drives the steering wheel, accelerator, or brake of the host vehicle in response to the control signal from the vehicle control unit 25, and the route estimation process according to this embodiment ends.

[0065] [Effects of the first embodiment] As described above in detail, the lane estimation device 1 according to this embodiment detects lane markings located on the left and right of the vehicle, calculates the parallelism of the detected lane markings, and sets a parallelism threshold to a reference value if the distance from the vehicle to the intersection is equal to or greater than a predetermined determination distance. On the other hand, if the distance from the vehicle to the intersection is less than the determination distance, the parallelism threshold is set higher than the reference value. Then, if the calculated parallelism is higher than the set parallelism threshold, the lane markings are determined to be parallel. If the lane markings are determined to be parallel, the lane estimation device 1 estimates the lane on which the vehicle will travel based on the lane markings. This allows the device to quickly determine the increase in the number of lanes and estimate the appropriate lane, even in locations where the number of lanes increases before an intersection.

[0066] In particular, the lane estimation device 1 according to this embodiment can strictly determine whether the left and right lane markings are parallel when the vehicle approaches an intersection, so that an increase in the number of lanes can be determined early even in a location where the number of lanes increases before the intersection. This prevents erroneous lane estimation that expands toward the increased lane even when the number of lanes increases before the intersection, making it possible to estimate the appropriate lane.

[0067] Furthermore, in the lane estimation device 1 according to this embodiment, if it is determined that the left and right lane marks are not parallel, the lane marking that is the smaller in difference from the direction of the lane the host vehicle is traveling in is set as the reference lane marking, and the lane on which the host vehicle will travel is estimated based on the reference lane marking. As a result, even if the left and right lane marks are not parallel, the lane marking that is the smaller in difference from the direction of the lane the host vehicle is traveling in is selected, and the lane marking on the side of the lane being added is not used, thereby preventing erroneous lane estimation that bulges out toward the lane being added.

[0068] Furthermore, in the lane estimation device 1 of this embodiment, if it is determined that the left and right dividing lines are not parallel, the reference dividing line is moved in the direction of the other of the left and right dividing lines by the lane width to set a virtual line, and the lane on which the vehicle will travel is estimated based on the virtual line and the reference dividing line. This means that when the left and right lane markings are not parallel, the lane marking that differs less from the direction of the lane the vehicle is traveling in is selected, and a virtual line is set from that lane marking, thereby preventing erroneous estimation of the route that bulges out toward the added lane.

[0069] Furthermore, the lane direction in which the vehicle is traveling is set based on the travel trajectory of the vehicle in the travel path estimation device 1 according to this embodiment. This allows the direction of the lane in which the vehicle is traveling to be accurately set when the left and right lane markings are not parallel, thereby preventing erroneous estimation of the lane direction, such as expanding toward the added lane, and enabling the appropriate lane to be estimated.

[0070] Furthermore, the lane estimation device 1 according to this embodiment sets the direction of the lane in which the vehicle is traveling based on the direction of the road the vehicle is scheduled to travel on. This allows the direction of the lane in which the vehicle is traveling to be accurately set when the left and right lane markings are not parallel, preventing erroneous estimation of the lane that expands toward the added lane and enabling the appropriate lane to be estimated.

[0071] Furthermore, the lane estimation device 1 according to this embodiment sets the direction of the lane in which the vehicle is traveling based on the direction of the lane adjacent to the lane the vehicle is traveling in. This allows the direction of the lane in which the vehicle is traveling to be accurately set when the left and right lane markings are not parallel, thereby preventing erroneous estimation of the lane, such as sagging toward the added lane, and enabling the appropriate lane to be estimated.

[0072] Furthermore, the lane estimation device 1 according to this embodiment sets the direction of the lane the vehicle is traveling in based on the arrow marking of the lane the vehicle is traveling in. This allows the direction of the lane the vehicle is traveling in to be accurately set when the left and right lane markings are not parallel, preventing erroneous estimation of the lane that expands toward the added lane and enabling the appropriate lane to be estimated.

[0073] Furthermore, the lane estimation device 1 according to this embodiment sets a switching distance shorter than the judgment distance, and switches the parallelism threshold from a value higher than the reference value to the reference value when the distance from the vehicle to the intersection is less than the switching distance. This increases the parallelism threshold only in the section where the number of lanes increases before the intersection, making it possible to strictly judge whether the lane lines are parallel only in the necessary sections, and efficiently determining whether the left and right lane lines are parallel.

[0074] Furthermore, in the lane estimation device 1 according to this embodiment, if a new lane marking is detected after the distance from the vehicle to the intersection falls below the judgment distance and it has been determined that the left and right lane markings are not parallel, the parallelism threshold is switched from a value higher than the reference value to the reference value. As a result, if a new lane marking is detected before the intersection, the parallelism threshold is returned to the reference value, eliminating the needlessly strict judgment of parallelism and enabling the parallelism of the left and right lane markings to be determined efficiently.

[0075] Furthermore, in the lane estimation device 1 according to this embodiment, the parallelism threshold is set higher than the reference value only when the intersection is an intersection where a right or left turn is possible. This prevents unnecessary strictness in determining whether the intersection is parallel at intersections where a right or left turn is not possible, and makes it possible to efficiently determine whether the left and right lane markings are parallel.

[0076] [Second embodiment] A second embodiment of the present invention will be described below with reference to the drawings. In the drawings, the same parts are denoted by the same reference numerals, and detailed description thereof will be omitted. The lane estimation device 1 according to this embodiment has the same configuration as the first embodiment shown in FIG.

[0077] Fig. 13 is a flowchart showing the processing steps of the path estimation processing by the path estimation device 1 according to this embodiment. As shown in Fig. 13, the path estimation processing according to this embodiment differs from the path estimation processing of the first embodiment shown in Fig. 5 in that steps S201 to S205 are added.

[0078] The processing of steps S101 to S117 is executed in the same manner as in the first embodiment, and in step S117, the parallel determination unit 21 determines whether the parallelism calculated in step S115 is higher than the parallelism threshold set in step S109 or step S113. If the parallelism is higher than the parallelism threshold, it is determined that the lane markings located on the left and right of the host vehicle are parallel, and the process proceeds to step S119, but if the parallelism is equal to or less than the parallelism threshold, it is determined that the lane markings located on the left and right of the host vehicle are not parallel, and the process proceeds to step S201.

[0079] In step S201, the lane estimation unit 23 sets a predetermined area between the lane markings located on the left and right of the host vehicle. As shown in Fig. 14, the lane estimation unit 23 sets the lane marking L1, which is the smaller of the lane markings L1 and L2 located on the left and right of the host vehicle, as the reference lane marking. Then, the lane estimation unit 23 sets a predetermined area 40 having a predetermined width at a position where the reference lane marking L1 is shifted in the direction of the lane marking L2 by the lane width.

[0080] In step S203, the lane estimation unit 23 determines whether or not a road marking indicating a lane marking exists within the predetermined area 40 set in step S201. If a road marking indicating a lane marking exists within the predetermined area 40, the process proceeds to step S205, and if a road marking indicating a lane marking does not exist within the predetermined area 40, the process proceeds to step S121.

[0081] In step S205, the roadway estimation unit 23 estimates the roadway on which the host vehicle will travel based on the lane markings that existed within the predetermined area 40 and the reference lane markings. As shown in Fig. 14, the roadway estimation unit 23 estimates the lane marking L1 set as the reference lane marking and the lane markings that exist within the predetermined area 40 as the roadway on which the host vehicle will travel.

[0082] Thereafter, similarly to the first embodiment, the processes of steps S119 to S123 are executed, and the lane estimation process according to this embodiment ends.

[0083] [Effects of the second embodiment] As described above in detail, in the lane estimation device 1 according to this embodiment, if it is determined that the left and right lane markings are not parallel, the reference lane marking is moved in the direction of the other lane marking by the lane width to set a predetermined area. It then determines whether a road marking indicating a lane marking is present within the predetermined area, and if so, it estimates the lane on which the vehicle will travel based on the lane marking present within the predetermined area and the reference lane marking. As a result, if the left and right lane markings are not parallel, the lane marking that differs least from the direction of the lane on which the vehicle is traveling is selected, and the predetermined area is set from that lane marking to detect the other lane marking. This prevents erroneous lane estimation, such as extending toward the added lane.

[0084] The above-described embodiment is merely an example of the present invention, and therefore the present invention is not limited to the above-described embodiment, and various modifications can be made to the design and other aspects of the present invention without departing from the technical concept of the present invention. [Explanation of symbols]

[0085] 1 Track estimation device 3. Map information receiver 5 sensors 7 GNSS receiver 9 Actuators 11 Vehicle position estimation unit 13 Intersection information acquisition unit 15. Lane line detection unit 17 Parallelism calculation section 19 Parallelism threshold setting section 21 Parallel Judgment Section 23 Route estimation part 25 Vehicle control unit 30, 32 Arrow signs 40 Predetermined area 100 Vehicle Systems

Claims

1. Obtaining a distance from the vehicle to an intersection ahead on a road on which the vehicle is traveling; Detecting lane markings located on the left and right sides of the vehicle; Calculating a parallelism indicating a degree of parallelism between the detected left and right lane markings; When the distance from the host vehicle to the intersection is equal to or greater than a predetermined determination distance, a parallelism threshold for determining whether or not the lane markings located on the left and right of the host vehicle are parallel is set to a reference value, and when the distance from the host vehicle to the intersection is less than the determination distance, the parallelism threshold is set to a value higher than the reference value; If the calculated parallelism is higher than the set parallelism threshold, it is determined that the left and right lane markings are parallel; When it is determined that the left and right lane markings are parallel, the route on which the host vehicle will travel is estimated based on the left and right lane markings. A method for estimating a running path.

2. The lane estimation method described in claim 1, characterized in that if it is determined that the left and right lane markings are not parallel, the lane marking that is less different from the direction of the lane in which the vehicle is traveling is set as the reference lane marking, and the lane on which the vehicle is traveling is estimated based on the reference lane marking.

3. 3. The lane estimation method according to claim 2, wherein, when it is determined that the left and right lane markings are not parallel, the reference lane marking is moved in the direction of the other lane marking to set a predetermined area, and it is determined whether or not a road marking indicating a lane marking exists within the predetermined area. If a road marking exists, the lane marking that the vehicle will travel on is estimated based on the lane marking that exists within the predetermined area and the reference lane marking.

4. The lane estimation method described in claim 2, characterized in that if it is determined that the left and right dividing lines are not parallel, the reference dividing line is moved in the direction of the other of the left and right dividing lines by the width of the lane to set a virtual line, and the lane on which the vehicle will travel is estimated based on the virtual line and the reference dividing line.

5. 3. The method for estimating a travel path according to claim 2, wherein the direction of the lane in which the host vehicle is traveling is set based on the travel path of the host vehicle.

6. 3. The method for estimating a travel path according to claim 2, wherein the direction of the lane in which the host vehicle is traveling is set based on the direction of a road on which the host vehicle is scheduled to travel.

7. 3. The lane estimating method according to claim 2, wherein the direction of the lane in which the host vehicle is traveling is set based on the direction of a lane adjacent to the lane in which the host vehicle is traveling.

8. 3. The lane estimating method according to claim 2, wherein the direction of the lane in which the host vehicle is traveling is set based on an arrow marking of the lane in which the host vehicle is traveling.

9. A switching distance shorter than the judgment distance is set, and when the distance from the vehicle to the intersection is less than the switching distance, the parallelism threshold is switched from a value higher than the reference value to the reference value. A lane estimation method according to any one of claims 1 to 8.

10. A lane estimation method according to any one of claims 1 to 8, characterized in that when a new lane line is detected after the distance from the vehicle to the intersection becomes less than the judgment distance and it is determined that the left and right lane lines are not parallel, the parallelism threshold is switched from a value higher than the reference value to the reference value.

11. The lane estimation method according to any one of claims 1 to 8, characterized in that the parallelism threshold is set higher than the reference value only when the intersection is an intersection where a right turn or a left turn is possible.

12. A path estimation device including a controller that estimates a path on which a vehicle travels, The controller Acquire a distance from the vehicle to an intersection ahead on a road on which the vehicle is traveling; Detecting lane markings located on the left and right sides of the vehicle; Calculating a parallelism indicating a degree of parallelism between the detected left and right lane markings; When the distance from the host vehicle to the intersection is equal to or greater than a predetermined determination distance, a parallelism threshold for determining whether or not the lane markings located on the left and right of the host vehicle are parallel is set to a reference value, and when the distance from the host vehicle to the intersection is less than the determination distance, the parallelism threshold is set to a value higher than the reference value; If the calculated parallelism is higher than the set parallelism threshold, it is determined that the left and right lane markings are parallel; When it is determined that the left and right lane markings are parallel, the route on which the host vehicle will travel is estimated based on the left and right lane markings. A path estimation device characterized by:

Images