Driving control method and driving control device

Abstract

The present invention provides a driving control method and a driving control device capable of estimating the lane in a wide area in front of the vehicle. [Solution] The controller acquires multiple feature points of objects around the vehicle detected by the surrounding detection sensor, identifies a first feature point indicating a lane boundary on the road from these feature points, acquires a reference line 31 along the target driving route on the road corresponding to the vehicle's current position from map information, transforms and superimposes the first feature points 32A onto the coordinate system of the reference line 31, extends the point cloud of these first feature points 32A in the direction of the path length along the reference line 31 to form a virtual lane boundary 33B, estimates the lanes on the road based on the lane boundary 33A based on the first feature points and the virtual lane boundary 33B, and sets a lane change section 34 based on the estimated lanes.

Description

Technical Field

[0001] The present invention relates to a driving control method and a driving control device.

Background Art

[0002] In vehicle automatic driving technology and driving support technology, when a host vehicle changes lanes, it is necessary to appropriately recognize the adjacent lane. Even when the lane marking of the adjacent lane is blocked, a technology for estimating the lane marking of the adjacent lane is known (see, for example, Patent Document 1). In the technology described in Patent Document 1, within the range recognized by an image recognition device, a lane marker (lane marking) of the host lane in which the host vehicle travels, a lane marking of an adjacent lane adjacent to the host lane, and the lane width of the adjacent lane are detected. Then, a position shifted by the detected lane width of the adjacent lane from the lane marking of the host lane in front of the host vehicle is estimated as the position of the lane marking of the adjacent lane. Thereby, even when the lane marking of the adjacent lane is blocked by another vehicle or the like, the adjacent lane to which the lane change destination is located can be estimated.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the technology described in Patent Document 1 has a problem that the estimated position of the lane marking (lane boundary line) of each lane is limited to the distance range detectable by the image recognition device. In such a case, for example, in a scene where the host vehicle is driven by automatic driving, a lane change section in which the host vehicle changes lanes from the host lane to an adjacent lane is set, and the lane change is executed in the lane change section, the lane change section can be set only within a limited range, and the lane change section becomes short.

[0005] The present invention aims to provide a driving control method and a driving control device that can estimate the lane in a wide area in front of the vehicle. [Means for solving the problem]

[0006] A driving control method according to a first aspect of this disclosure acquires multiple feature points of objects around the vehicle detected by an ambient detection sensor, and identifies a first feature point indicating a lane boundary on the road from these feature points. It also acquires a reference line along the target driving route on the road corresponding to the vehicle's current position from map information having a ground coordinate system. The first feature points are transformed and superimposed on the ground coordinate system of this reference line, and the point cloud of these first feature points is extended in the direction of the path length along the reference line to form a virtual lane boundary, and the lane on the road is estimated based on the lane boundary based on the first feature points and the virtual lane boundary. Then, a lane change section is set based on the estimated lane, in which the vehicle changes lanes from its own lane to an adjacent lane.

[0007] This allows the system to estimate lane markings even when they are obscured by other vehicles or other obstacles. Based on these estimated markings, the system can estimate the lanes on the road over a wide area beyond the detection range of the surrounding sensors. Therefore, it is possible to appropriately set lane change sections when configuring them. [Brief explanation of the drawing]

[0008] [Figure 1] A block diagram showing the schematic configuration of a vehicle according to one embodiment of the present disclosure. [Figure 2] This figure illustrates the method for estimating the alignment of a vehicle according to this embodiment. [Figure 3] This figure shows an example of a lane change section set up in this embodiment. [Figure 4] A flowchart illustrating the operation control method of this embodiment. [Modes for carrying out the invention]

[0009] An embodiment of this disclosure will be described below. Figure 1 is a block diagram showing the schematic configuration of the vehicle (Vehicle 1) of this embodiment. The vehicle 1 in this embodiment is a vehicle equipped with the driving control device of this disclosure, and as shown in Figure 1, it includes an ambient detection sensor 11, a current position detection sensor 12, a driving state detection sensor 13, a navigation device 14, an actuator 15, a controller 20, and the like. The surrounding detection sensor 11 is a sensor that detects objects in the vicinity of the vehicle 1. Examples of the surrounding detection sensor 11 include a camera that captures images of the surroundings, a laser radar, a millimeter-wave radar, an ultrasonic radar, a laser rangefinder, etc., and a combination of these may be used. The surrounding objects detected by the surrounding detection sensor 11 include road conditions for the vehicle 1 such as road lanes, traffic signs, and signals, and obstacles around the vehicle 1 (other vehicles, pedestrians, etc.). The surrounding detection sensor 11 can also measure the distance of the detected surrounding objects from the vehicle 1.

[0010] The current position detection sensor 12 is a sensor that detects the current position of the vehicle. Examples of the current position detection sensor 12 include a receiver that receives satellite signals from a GNSS (Global Navigation Satellite System) to determine the current position.

[0011] The driving state detection sensor 13 is a group of various sensors that detect the driving state of the vehicle 1. Examples of the driving state detection sensor 13 include a wheel speed sensor, a gyro sensor, an acceleration sensor, and a magnetic sensor. The wheel speed sensor measures the wheel speed of the vehicle 1. The gyro sensor is a sensor that measures the angular velocity of the vehicle 1 and may include a yaw angle sensor and a yaw rate sensor. The acceleration sensor measures the acceleration of the vehicle 1. The acceleration sensor may include a longitudinal acceleration sensor that measures the acceleration of the vehicle 1 in the longitudinal direction and a lateral acceleration sensor that measures the acceleration of the vehicle 1 in the lateral direction. The magnetic sensor measures the azimuth angle of the direction of travel of the vehicle 1. A configuration may be provided in which the azimuth angle can be calculated based on the measured values ​​of the gyro sensor (yaw angle sensor and yaw rate sensor) without a magnetic sensor.

[0012] The navigation device 14 searches for a target route to the destination of the vehicle 1 based on destination information entered by the user (passenger) and map information. The map information may be stored in the memory installed in the navigation device 14, stored in a separately provided map database, or obtained from an external device using a communication line such as the internet.

[0013] The map information of this embodiment includes at least road information that models the shape of a road by connecting multiple links and nodes. The map information is information that indicates each point of the links and nodes by absolute position (e.g., latitude and longitude). The coordinate system that indicates the position of each point based on such absolute position will hereafter be referred to as the ground coordinate system. In addition to the road information described above, map information also includes more detailed information such as the number of lanes and lane width of each road, but this detailed information is not essential. In this embodiment, even if the map information only contains road model information, the number of lanes and lane width on the road on which vehicle 1 is traveling are appropriately calculated.

[0014] The actuator 15 is a drive device that moves the vehicle based on the control of the controller 20, and includes, for example, an engine actuator, a brake hydraulic actuator, and a steering angle actuator. The engine actuator controls the engine's driving force based on drive control commands from the controller 20. In the case of a hybrid vehicle, both an engine actuator and a motor actuator may be used, and in the case of an electric vehicle, only a motor actuator may be used. The brake hydraulic actuator is a hydraulic booster that controls the brake hydraulic force based on braking control commands from the controller 20. In the case of electric vehicles that do not have a hydraulic booster, an electric booster may be used. The steering angle actuator is a steering angle control motor that controls the steering angle of the steering wheels based on a steering angle control command from the controller 20.

[0015] The controller 20 is a computer that controls the autonomous driving of the vehicle. The controller 20 is composed of a storage device such as semiconductor memory and an arithmetic circuit such as a CPU (Central Processing Unit). The controller 20 realizes various functions by having the arithmetic circuit read and execute programs stored in the storage device. Specifically, as shown in Figure 1, the controller 20 functions as an surrounding situation acquisition unit 21, a lane detection unit 22, a route acquisition unit 23, a reference line acquisition unit 24, a self-position update unit 25, a lane shape estimation unit 26, a lane change planning unit 27, and a driving control unit 28, etc. In this example, the arithmetic circuit of the controller 20 executes a program to realize the various functional configurations of the surrounding situation acquisition unit 21, lane detection unit 22, route acquisition unit 23, reference line acquisition unit 24, self-position update unit 25, lane shape estimation unit 26, lane change planning unit 27, and driving control unit 28. However, some or all of these may be realized by individual hardware configurations.

[0016] The surrounding situation acquisition unit 21 acquires the surrounding detection results detected by the surrounding detection sensor 11. Examples of the surrounding detection results include, for example, the surrounding captured image captured by a camera, the position and distance of the laser reflection points on the surrounding objects acquired by a lidar, etc.

[0017] The lane detection unit 22 detects the lanes within the detection range of the surrounding detection sensor 11 based on the captured image, laser reflection points, etc. acquired by the surrounding situation acquisition unit 21. That is, the feature points of the target object are acquired by the surrounding detection sensor 11, and the first feature points indicating the dividing lines, road edges, median strips, etc. that divide the lanes are extracted to detect the positions and the number of lanes of each lane on the road. Examples of the feature points acquired by the surrounding detection sensor 11 include, for example, the pixel positions where the luminance value changes greatly in the captured image captured by a camera, the laser reflection points, etc. By combining the captured image of the camera and the laser reflection points of the lidar, or using a stereo camera, the first feature points indicating the dividing lines, road edges, and median strips on the road, and the distance from the host vehicle 1 to the first feature points can be accurately detected.

[0018] The route acquisition unit 23 acquires the target driving route of the host vehicle 1 set by the navigation device 14. That is, based on the map information, the route along the road from the current position of the host vehicle 1 to the target point is acquired.

[0019] The reference line acquisition unit 24 acquires, as a reference line, the line along the road of the target driving route at the current position of the host vehicle 1. For example, in the target driving route, when the host vehicle 1 turns right at the next intersection on the road where it is currently driving, the line that goes straight until the intersection and turns right at the intersection to enter the right-turn destination road is the reference line. In the target driving route, when the host vehicle 1 goes straight at the intersection on the road where it is currently driving, the line along the current road (the line for going straight through the intersection) is the reference line.

[0020] The own - position updating unit 25 updates the current position of the host vehicle 1. The current position of the host vehicle 1 is updated with the positioning position detected by the current - position detection sensor 12 as the current position of the host vehicle 1 under the situation where the host vehicle 1 is not changing lanes, that is, when the host vehicle 1 is going straight along its own lane. On the other hand, under the situation where the host vehicle 1 is changing lanes, the current position is updated based on odometry. That is, when the host vehicle 1 moves from the current position to the target lane, the controller 20 needs to recognize the current position of the host vehicle 1 with high precision. However, there may be a large error in the current position detected by the current - position detection sensor 12. Therefore, based on the vehicle speed, the longitudinal acceleration, the lateral acceleration, and the azimuth angle of the host vehicle 1 measured by the driving - state detection sensor 13, the position of the host vehicle 1 on the road is calculated. In addition, the current position of the host vehicle 1 may be updated by correcting the current position based on the current - position detection sensor 12 with odometry.

[0021] The lane - shape estimation unit 26 estimates the shape of the lane on the road by using the feature points for determining the lane detected by the lane - detection unit 22 and the reference line acquired by the reference - line acquisition unit 24. FIG. 2 is a diagram showing the method for estimating the lane shape of the present embodiment. Here, the reference line 31 is a line along the target driving route in the map information, and is information indicating the absolute position (for example, longitude and latitude) of each point of the node and the link in the ground coordinate system. On the other hand, the first feature points indicating lane boundaries (lane markings, road edges, median strips, etc.) acquired by the lane detection unit 22 are in a vehicle coordinate system acquired with reference to the surrounding detection sensor 11, and can be represented by the distance s to the first feature point along the path length of the vehicle 1 (forward in the direction of travel) and the distance d(s) to the first feature point in a direction perpendicular to the path length. The lane shape estimation unit 26 converts these vehicle coordinate system coordinate points (s, d(s)) of the first feature points into ground coordinate system coordinates (X(s), Y(s)) by multiplying them by a predetermined conversion function f(s, d(s)). Since the above coordinate system conversion is performed with the position relative to the path length direction (distance s from the vehicle 1) as a common parameter, the inverse conversion from the ground coordinate system to the vehicle coordinate system can also be easily performed using the conversion function f(s, d(s)), so that the ground coordinate system coordinates of the absolute positions of each point in a node or link can be converted into vehicle coordinate system coordinate points. By performing one of the coordinate system transformations described above, the first feature point 32A and the reference line 31 can be compared on the same coordinate system. Each first feature point 32A is obtained as a point cloud indicating the position of the lane boundary line 33A, as described above. The lane shape estimation unit 26 estimates a virtual lane boundary line 33B outside the detection range of the surrounding detection sensor 11 by extending the lane boundary line 33A, which is identified by the point cloud of first feature points 32A, parallel to the reference line 31. Note that 32B in Figure 2 indicates a virtual feature point on the virtual lane boundary line 33B. The region enclosed by these lane boundary lines 33A or virtual lane boundary lines 33B is then estimated as the lane on the road (own lane 35, adjacent lane 36). This makes it possible to understand the lane shape in front of the own vehicle 1 even when the dividing line cannot be detected by a preceding vehicle 2, etc., as shown in Figure 2. It also makes it possible to estimate lanes outside the detection range of the surrounding detection sensor 11.

[0022] The lane change planning unit 27 determines whether or not to change the lane of vehicle 1 based on the direction of travel at the intersection on the target driving path of vehicle 1. If the lane change planning unit 27 determines that a lane change is to be made, it sets a lane change section in front of vehicle 1 up to the intersection. For example, if vehicle 1 is turning right at the intersection, it determines that vehicle 1 should change lanes to the rightmost lane on the road, and if vehicle 1 is turning left at the intersection, it determines that vehicle 1 should change lanes to the leftmost lane on the road. If vehicle 1 is going straight at the intersection, it determines that lanes should be changed to the center lane or the left lane (in countries with left-hand traffic). In this case, the unit may further determine whether or not a lane change is possible after a road sign indicating a lane where going straight is permitted is detected by the surrounding detection sensor 11.

[0023] Figure 3 shows an example of a lane change section set up in this embodiment. The lane change planning unit 27 sets a lane change section 34 in front of the vehicle 1 and before the lane change prohibited section (for example, the section 30m from the intersection) which is set before the intersection, and the section is a predetermined distance in the direction of the route length. Specifically, the lane change planning unit 27 sets the starting point 34A and ending point 34B when performing a lane change in the lane change section 34, and generates a target trajectory 34C connecting these starting point 34A and ending point 34B. In this embodiment, the current driving line 35A is defined as a line parallel to the reference line 31 and passing over the vehicle 1 in the vehicle's own lane 35 on which the vehicle 1 is traveling. In other words, the current driving line 35A is defined as a line parallel to the reference line 31 from the current position of the vehicle 1 in the lateral direction perpendicular to the path length direction. The target driving line 36A is defined as a line that passes through the center of the adjacent lane 36 to which the vehicle 1 is changing lanes and is parallel to the reference line 31. The lateral distance from the current driving line 35A to the target driving line 36A corresponds to the distance between the vehicle's own lane 35 and the adjacent lane 36 in this disclosure. The lane change planning unit 27 sets the starting point 34A on the current driving line 35A, which is on the extension of the vehicle 1's current position in the lateral direction. The lane change planning unit 27 also sets the ending point 34B on the target driving line 36A, which is moved laterally from the current driving line 35A by the distance between the vehicle's own lane 35 and the adjacent lane 36.

[0024] The lane change planning unit 27 sets a starting point 34A in the direction of the route length, at a position that is a distance d1 ahead of the vehicle 1 calculated based on the vehicle's current speed and the turn signal activation time. The turn signal activation time can be a preset fixed value, for example, set to 3 seconds. If the vehicle's current speed is faster than a predetermined speed, the lane change planning unit 27 may also create a speed profile that decelerates the vehicle 1 so that the lateral acceleration is below a threshold before changing lanes. In this case, the unit calculates the position where deceleration is completed from the vehicle 1, and sets the starting point 34A at a position that is a distance ahead of the position where deceleration is completed, based on the vehicle's speed after deceleration and the turn signal activation time. Furthermore, the lane change planning unit 27 sets the tangential direction of the current lane 35A at the starting point 34A as the azimuth angle of the vehicle 1 at the starting point 34A.

[0025] The lane change planning unit 27 sets the end point 34B at a distance d2 ahead of the intersection point P of the normal vector at the starting point 34A of the current lane 35A and the target lane 36A, based on the time required for the lane change and the speed of the vehicle 1. The time required for the lane change can be a preset fixed value, for example, set to a value between 5 and 10 seconds. The speed of the vehicle 1 can be exemplified by the speed of the vehicle 1 at the starting point 34A. As mentioned above, if the vehicle 1 is decelerated in advance before changing lanes, the speed after deceleration may be used. Furthermore, the lane change planning unit 27 sets the tangential direction of the target travel line 36A at the end point 34B as the azimuth angle of the vehicle 1 at the end point 34B.

[0026] The lane change planning unit 27 then connects the starting point 34A to the ending point 34B with a smooth curve to form the target trajectory 34C. For example, a cubic function including the starting point 34A and the ending point 34B is derived, and the target trajectory 34C is generated using the curve of that cubic function. However, the method of generating the target trajectory is not limited to this, and it is sufficient if the starting point 34A and the ending point 34B are connected by a smooth curve with an inflection point.

[0027] The driving control unit 28 performs automatic driving control of the vehicle 1. For example, the driving control unit 28 controls the actuator 15 so that the vehicle 1 travels along the current driving line 35A in its own lane at a constant speed, or follows a preceding vehicle, thereby controlling the vehicle's movement. Furthermore, when a lane change section 34 is set by the lane change planning unit 27, the driving control unit 28 controls the actuator 15 to follow the generated target trajectory 34C.

[0028] [Vehicle control method] Next, the driving control method of this embodiment will be described. Figure 4 is a flowchart relating to the driving control method of this embodiment. In this embodiment, for example, when the navigation device 14 initiates the automatic driving process of the vehicle 1 based on the target driving route to the destination, the driving control unit 28 of the controller 20 outputs control commands to the actuator 15 regarding the driving of the vehicle 1 along the target driving route, causing the vehicle to drive. In the above automated driving process, the surrounding situation acquisition unit 21 acquires the surrounding detection results of the vehicle 1 based on the sensor signals input from the surrounding detection sensor 11, and the lane detection unit 22 detects the first characteristic point 32A of the lane boundary line (road edge, median strip, etc.) on the road based on the surrounding detection results (step S1). Furthermore, the route acquisition unit 23 acquires the target driving route set by the navigation device 14 (step S2), and the reference line acquisition unit 24 acquires a reference line 31 along the target driving route (step S3).

[0029] Then, the lane shape estimation unit 26 converts the first feature point 32A of the vehicle coordinate system acquired in step S1 into the same ground coordinate system as the reference line 31, or converts the reference line 31 of the ground coordinate system into the same vehicle coordinate system as the first feature point 32A, thereby converting the first feature point 32A and the reference line 31 into the same coordinate system (step S4). The lane shape estimation unit 26 also estimates a virtual lane boundary line 33B by extending the lane boundary line 33A, which can be identified from the point cloud data of the first feature point 32A after the coordinate transformation, so that it is parallel to the reference line 31 (step S5). Then, based on the lane boundary line 33A and the virtual lane boundary line 33B, the shape of each lane on the road (own lane 35, adjacent lane 36, etc.) is estimated (step S6). This makes it possible to estimate the position and shape of each lane outside the detection range of the surrounding detection sensor 11.

[0030] Furthermore, the self-position update unit 25 acquires and updates the current position of the vehicle 1 while the vehicle 1 is in motion (step S7). Note that the current position update is performed continuously while the vehicle 1 is in motion. As described above, if the vehicle 1 is not performing a lane change, the self-position update unit 25 updates the current position based on the positioning position detected by the current position detection sensor 12, and if the vehicle 1 is performing a lane change, it updates the current position based on odometry. Therefore, in step S7, the self-position update unit 25 updates the current position using the positioning position detected by the current position detection sensor 12.

[0031] The lane change planning unit 27 determines the direction of travel for vehicle 1 at the next intersection ahead of vehicle 1, based on the vehicle's current position, map information, and target travel route (step S8). The lane change planning unit 27 then determines whether a lane change is necessary for the intersection (step S9). For example, if the next intersection on the target route requires a right turn, the unit determines that a lane change is not necessary if vehicle 1 is traveling in the rightmost lane, and if it is traveling in any other lane, it determines that a lane change to the right lane is necessary. The same applies to left turns; if vehicle 1 is traveling in the leftmost lane, it determines that a lane change is not necessary, and if it is traveling in any other lane, it determines that a lane change to the left lane is necessary. If the next intersection on the target route requires going straight, the unit determines that a lane change is not necessary if vehicle 1 is traveling in the center lane of a road with three or more lanes on one side, and if it is traveling in any other lane, it determines that a lane change to the center lane is necessary. Alternatively, the system may determine whether the vehicle is traveling in a lane where it can go straight at an intersection based on whether the country in which the vehicle 1 is traveling has left-hand or right-hand traffic. For intersections where right turns are prohibited, the right lane may be considered a lane where the vehicle can go straight, and for intersections where left turns are prohibited, the left lane may be considered a lane where the vehicle can go straight, and the system may determine whether the vehicle's lane 35 can go straight. Alternatively, the system may determine whether a lane change is necessary when a road sign is recognized by the surrounding detection sensor 11.

[0032] If the result in step S9 is determined to be YES, the lane change planning unit 27 sets the lane change section 34 (step S10). Specifically, the lane change planning unit 27 sets a starting point 34A on the current driving lane 35A at a distance d1 in front of the vehicle 1 based on the current speed and the duration of the turn signal operation. From the intersection point P of the normal vector at the starting point 34A of the current driving lane 35A and the target driving lane 36A, the ending point 34B is set at a distance d2 in front of the intersection point P, based on the time required for the lane change and the speed of the vehicle 1 at the time of the lane change. The lane change planning unit 27 also calculates the target trajectory 34C by deriving a cubic function connecting the starting point 34A and the ending point 34B, using the azimuth angle at the starting point 34A as the tangential direction to the current driving lane 35A and the azimuth angle at the ending point 34B as the tangential direction to the target driving lane 36A.

[0033] After step S10, the lane change planning unit 27 outputs a lane change instruction for the lane change section 34 to the driving control unit 28. As a result, the driving control unit 28 controls the actuator 15 to perform a lane change along the generated target trajectory 34C (step S11). For example, the driving control unit 28 converts the current driving line 35A and target driving line 36A of the vehicle 1 estimated in step S6, and the lane change section 34 (target trajectory 34C from the start point 34A to the end point 34B) set in step S10 from the ground coordinate system to a vehicle coordinate system centered on the vehicle 1, and outputs command values ​​to the actuator 15 to drive the vehicle 1 along the converted current driving line 35A, target trajectory 34C, and target driving line 36A. Note that if the reference line 31 of the ground coordinate system was converted to the same vehicle coordinate system as the first feature point 32A in step S4, it is not necessary to convert the lane change section 34 from the ground coordinate system to a vehicle coordinate system centered on the vehicle 1.

[0034] Furthermore, as described above, the self-position update unit 25 updates its current position based on odometry while the lane change in step S11 is being performed (step S12). This allows for accurate determination of the position of the vehicle 1 during the lane change, and enables appropriate automated driving of the vehicle 1 based on position information.

[0035] On the other hand, if it is determined in step S9 that NO (no lane change will be performed), and after the lane change in steps S11 to S12 is completed, the driving control unit 28 performs follow control of the vehicle 1 using normal automatic driving operation based on each lane on the road estimated in step S6, map information, the current position of the vehicle 1 updated by the self-position update unit 25, and the surrounding conditions detected by the surrounding detection sensor 11 (for example, obstacles such as preceding vehicles, adjacent vehicles, and pedestrians) (step S13). In step S13, the driving control unit 28 will, for example, make the vehicle 1 drive by following a preceding vehicle, or make the vehicle 1 drive by following the lane it is currently driving in (such as the current driving lane 35A or the target driving lane 36A). Note that in the flowchart shown in Figure 4, the process ends after step S13, but in reality, the process returns to step S1 and the above process is repeated.

[0036] [Effects of this embodiment] The vehicle (own vehicle 1) of this embodiment includes an ambient detection sensor 11 and a controller 20. The controller 20 identifies a first feature point 32A indicating a lane marking on the road from a plurality of feature points detected by the ambient detection sensor 11. It also obtains a reference line 31 from map information that follows the target driving route of the own vehicle 1 on the road corresponding to the current position of the own vehicle 1. Then, it estimates the lanes on the road from a virtual lane boundary line 33B obtained by extending the point cloud of these first feature points 32A in the direction of the path length along the reference line 31. Based on the estimated lanes, it sets a lane change section 34 in which the own vehicle 1 changes lanes from its own lane 35 to an adjacent lane 36.

[0037] As a result, in this embodiment, even when the lane boundary line is not within the sensor range of the surrounding detection sensor 11 due to obstruction by an obstacle such as a preceding vehicle, or when the sensor range is narrowed due to bad weather, it is possible to estimate lanes outside the sensor range. In addition, in this embodiment, it is also possible to estimate lanes further ahead of the effective sensor range of the surrounding detection sensor 11. Thus, in this embodiment, it is possible to estimate lanes over a wide area in front of the vehicle 1, and based on this, the position of the lane change section 34 when the vehicle 1 changes lanes can be appropriately set.

[0038] In this embodiment, the controller 20 sets the start point 34A, end point 34B, and target trajectory 34C of the lane change section 34 based on the current lateral position of the vehicle 1 and the distance between the vehicle's lane 35 and the adjacent lane 36. This allows for the setting of an appropriate target trajectory 34C in the lane change section, moving from the current lane 35A on which vehicle 1 is traveling to the target lane 36A of the adjacent lane 36.

[0039] In this embodiment, the controller 20 sets the position of the starting point 34A in the lateral direction perpendicular to the reference line 31 on the current travel line 35A, and sets the azimuth angle of the vehicle at the starting point 34A to the tangential direction of the current travel line 35A at the starting point 34A. This allows the system to set a target trajectory 34C in which the vehicle 1 travels along the current lane 35A to the starting point 34A, and then gradually changes lanes to the adjacent lane 36 from the starting point 34A.

[0040] In this embodiment, the controller 20 sets the position of the starting point 34A in the path length direction to a forward position on the current travel line 35A, which is calculated based on the current speed of the vehicle 1 and the operating time of the turn signal. This allows for setting a target trajectory 34C for lane changes after a predetermined time has elapsed since the turn signal was activated. In other words, it enables proper lane changes in accordance with traffic rules.

[0041] In this embodiment, the controller 20 sets the position of the endpoint 34B based on the time required for the vehicle 1 to change lanes and the speed of the vehicle 1 at the time of the lane change. This allows the endpoint 34B to be set in a position where it can be entered without difficulty.

[0042] In this embodiment, the controller 20 sets an endpoint 34B on the target travel line 36A, and sets the azimuth angle of the vehicle 1 at the endpoint 34B to the tangential direction at the endpoint 34B on the target travel line 36A. This allows setting a target trajectory 34C that gradually changes lanes from the endpoint 34B of the target trajectory 34C to the target running line 36A of the adjacent lane 36.

[0043] In this embodiment, the target trajectory 34C is set to be a trajectory that connects the starting point 34A to the ending point 34B with a cubic function curve. As a result, compared to setting a target trajectory 34C that connects the starting point 34A to the ending point 34B in a straight line, for example, it becomes possible to make smooth lane changes without sudden steering or the resulting abnormal lateral acceleration, and to perform appropriate lane changes without causing discomfort to the occupants of vehicle 1 or surrounding vehicles.

[0044] In this embodiment, the controller 20 updates the current position of its own vehicle 1 using odometry while traveling through the lane change section 34. In detecting the current position using the current position detection sensor 12, positioning errors may occur. If an error occurs in the current position of the vehicle 1 during a lane change, it may hinder a proper lane change along the target trajectory 34C. In contrast, in this embodiment, the current position is updated using odometry, that is, the position of the vehicle 1 is calculated from the driving state of the vehicle 1 detected by the driving state detection sensor 13. This suppresses the occurrence of such errors and enables a proper lane change along the target trajectory 34C.

[0045] [Differentiation] The present invention is not limited to the embodiments described above, but also includes the following modifications to the extent that the objectives of the present invention can be achieved. [Example 1] In the above embodiment, an example was shown in step S8 for generating a target trajectory 34C for the lane change section 34 of the vehicle 1 from the starting point 34A to the ending point 34B. In contrast, a speed profile may also be generated for a series of paths in which the vehicle 1 travels from its current position, passing through the starting point 34A, along the target trajectory 34C to the ending point 34B, and then along the target travel line 36A. In this case, the position of at least one of the starting point 34A and the ending point 34B in the direction of the path length may be corrected based on the speed profile.

[0046] [Differentiation 2] In the above embodiment, the self-position update unit 25 updates the current position of the vehicle 1 based on odometry when changing lanes, and updates the current position based on the position determined by the current position detection sensor 12 in other cases, but is not limited to this. For example, in scenes requiring complex steering processing, such as when the vehicle 1 turns right or left at an intersection, or when the vehicle enters a branching lane at a branching point, the self-position update unit 25 may update the current position based on odometry.

[0047] Furthermore, updating the current location based on odometry is not mandatory. For example, if the accuracy of the current location detection sensor 12 is sufficiently high, the current location may be updated based on the positioning measured by the current location detection sensor 12.

[0048] [Difference 3] In the above embodiment, the estimated centerline of the adjacent lane 36 was used as the target driving line 36A, but the embodiment is not limited to this. For example, the target driving line 36A may be a line obtained by laterally shifting the current driving line 35A by the distance between the lane boundary lines 33A (lane width). In this case, if the current driving line 35A is to the left of the center of the current lane 35, the target driving line 36A will also be set to the left of the center of the adjacent lane 36.

[0049] [Differentiation Example 4] In the above embodiment, the starting point 34A in the path length direction was set to a position forward by a distance d1 calculated based on the speed of the vehicle 1 and the operating time of the turn signal, but it is not limited to this. As mentioned above, the speed of the vehicle 1 is not limited to the current speed of the vehicle 1, but may be a predetermined lane-changing speed achieved by decelerating in order to change lanes. In this case, the starting point 34A can be set by adding the distance based on the lane-changing speed and the operating time of the turn signal to the distance it takes for the vehicle to decelerate from its current speed to the lane-changing speed.

[0050] [Difference 5] Furthermore, while the distance d1 for setting the starting point 34A is calculated based on the speed of the vehicle 1 and the duration of the turn signal operation, it may also be a predetermined distance. In this case, it is not necessary to calculate the distance to the starting point 34A based on the duration of the turn signal operation and the speed of the vehicle 1, and the starting point 34A can be set using a predetermined distance, simplifying the process. Also, the predetermined distance may be changed according to the current speed of the vehicle 1. For example, if the vehicle 1 is traveling at the first speed, the starting point 34A is set at a position one distance ahead of the vehicle 1, and if the vehicle 1 is traveling at the second speed, which is faster than the first speed, the starting point 34A is set at a position two distances ahead of the vehicle 1, which is longer than the first distance. The same applies to the endpoint 34B. In the above embodiment, the distance d2 from point P to endpoint 34B was calculated based on the speed of the vehicle 1 and the time required for the lane change, but for example, the distance d2 could be a predetermined distance in the direction of the path length.

[0051] [Modification 6] In the above embodiment, the target trajectory 34C was set by a cubic function curve connecting the starting point 34A and the ending point 34B, but it is not limited to this. The target trajectory 34C may be any function that connects the starting point 34A to the ending point 34B with a smooth curve, such as a trigonometric function. [Explanation of Symbols]

[0052] 1...Own vehicle, 11...Surroundings detection sensor, 12...Current position detection sensor, 13...Driving state detection sensor, 14...Navigation device, 15...Actuator, 20...Controller, 21...Surroundings acquisition unit, 22...Lane detection unit, 23...Route acquisition unit, 24...Reference line acquisition unit, 25...Self-position update unit, 26...Lane shape estimation unit, 27...Lane change planning unit, 28...Driving control unit, 31...Reference line, 32A...First feature point, 33A...Lane boundary line, 33B...Virtual lane boundary line, 34...Lane change section, 34A...Start point, 34B...End point, 34C...Target trajectory, 35...Own lane, 35A...Current driving line, 36...Adjacent lane, 36A...Target driving line.

Claims

1. A driving control method that controls the operation of a vehicle using a computer, Multiple feature points are acquired by the surrounding detection sensor that detects the surrounding conditions of the vehicle, From the multiple aforementioned feature points, a first feature point indicating the lane boundary on the road is identified, The system includes road information that models roads by connecting multiple links and nodes, and from map information that shows the positions of the links and nodes in a ground coordinate system based on absolute position on a map, it obtains a reference line along the target route of the vehicle. Based on a virtual lane boundary line obtained by extending the point cloud of the first feature points in the direction of the path length along the reference line, the lane of the road on which the vehicle is traveling is estimated. A driving control method that sets a lane change section in which the vehicle changes lanes from its own lane to an adjacent lane, based on the estimated lane.

2. In setting the lane change section, the system sets the starting point for when the vehicle changes lanes in the lane change section, the ending point for when the lane change is completed, and the target trajectory connecting the starting point and the ending point, based on the vehicle's current position in the lateral direction perpendicular to the vehicle's path length direction and the distance between the vehicle's lane and the adjacent lane. The operation control method according to claim 1.

3. The line parallel to the aforementioned reference line and passing through the current position of the vehicle is defined as the current travel line. The position of the starting point in the lateral direction perpendicular to the aforementioned reference line is set on the current travel line, The azimuth angle of the vehicle at the aforementioned starting point is set to the tangential direction of the current travel line. The operation control method according to claim 2.

4. The line parallel to the aforementioned reference line and passing through the current position of the vehicle is defined as the current travel line. The position of the starting point in the direction of the path length along the aforementioned reference line is set to the position in front of the vehicle on the current driving line, calculated based on the vehicle's current speed and the duration of the turn signal operation. The operation control method according to claim 2.

5. The position of the endpoint along the length of the path parallel to the aforementioned reference line is determined based on the time required for the vehicle to change lanes and the speed of the vehicle at the time of the lane change. The operation control method according to claim 2.

6. The target driving line is defined as a line parallel to the aforementioned reference line and passing through the center of the lane to which the vehicle will change lanes. The azimuth angle of the vehicle at the aforementioned endpoint is set to the tangential direction at the endpoint on the target travel line. The operation control method according to claim 2.

7. The aforementioned target trajectory is a cubic function that connects the starting point and the ending point in the lane change section. The operation control method according to claim 2.

8. While driving through the lane change section, the vehicle's current position is updated using odometry. The operation control method according to claim 1.

9. It is equipped with an ambient detection sensor that detects the surrounding conditions of the vehicle, and a controller. The aforementioned controller, Multiple feature points detected by the surrounding detection sensor are acquired, From the multiple aforementioned feature points, a first feature point indicating the lane boundary on the road is identified, The system includes road information that models roads by connecting multiple links and nodes, and from map information that shows the positions of the links and nodes in a ground coordinate system based on absolute position on a map, it obtains a reference line along the target route of the vehicle. Based on a virtual lane boundary line obtained by extending the point cloud of the first feature points in the direction of the path length along the reference line, the lane of the road on which the vehicle is traveling is estimated. A driving control device that sets a lane change section in which the vehicle changes lanes from its own lane to an adjacent lane, based on the estimated lane.

Images