Trajectory generation method and trajectory generation apparatus

Abstract

In a trajectory generation method and a trajectory generation device (20) according to the present invention, when a vehicle (V) is to enter an intersection (B1), a candidate lane (L1, L2, L3), which is a candidate for a lane for the vehicle (V) to travel when exiting the intersection (B1), is detected using detection results of a detection device for detecting lanes, and when there are a plurality of candidate lanes (L1, L2, L3), the accuracy with which the vehicle (V) can travel each of the candidate lanes (L1, L2, L3) using autonomous travel control is calculated, and an exit lane for the vehicle (V) to travel when exiting the intersection (B1) is selected from among the candidate lanes (L1, L2, L3) on the basis of the accuracies.

Description

Trajectory generation method and trajectory generation device

[0001] The present invention relates to a trajectory generation method and a trajectory generation device.

[0002] When a vehicle passes through an intersection and there are more lanes ahead of the intersection than there are before the intersection, a vehicle control device is known that selects the approach lane ahead of the intersection depending on the position of the lane the vehicle is traveling in (Patent Document 1).

[0003] JP 2016-224802 A

[0004] In the above-mentioned conventional technology, if the number of lanes beyond the intersection cannot be accurately recognized, the vehicle cannot properly enter the lane beyond the intersection, and cannot continue autonomous driving along the lanes.

[0005] The problem to be solved by the present invention is to provide a trajectory generation method and a trajectory generation device that can continue autonomous driving along a traffic lane after exiting an intersection.

[0006] The present invention solves the above problem by calculating the probability that the vehicle will be able to travel in the candidate lanes using autonomous driving control when a vehicle enters an intersection and multiple candidate lanes are detected as candidates for the lane the vehicle will travel in when exiting the intersection, and then selecting an exit lane from the candidate lanes based on the calculated probability.

[0007] According to the present invention, autonomous driving along the lane can be continued after exiting an intersection.

[0008] 1 is a block diagram showing an example of an embodiment of a driving assistance device including a trajectory generation device according to the present invention. FIG. 2 is a plan view showing an example of a driving scene in which driving assistance is performed by the driving assistance device of FIG. 1. FIG. 3 is a flowchart showing an example of a processing procedure for accuracy calculation by an accuracy calculation unit of FIG. 1. FIG. 4 is a plan view showing an example of a driving trajectory generated by the trajectory generation device of FIG. 1. FIG. 5 is a plan view showing another example of a driving trajectory generated by the trajectory generation device of FIG. 1. FIG. 6 is a flowchart showing an example of a processing procedure in the driving assistance device of FIG. 1. FIG. 7 is a flowchart showing another example of the processing procedure in the driving assistance device of FIG. 1 (part 1). FIG. 8 is a flowchart showing another example of the processing procedure in the driving assistance device of FIG. 1 (part 2).

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The following description is based on the assumption that vehicles are driven on the left side of the road in countries where left-hand traffic regulations apply. In countries where right-hand traffic regulations apply, the left and right in the following description should be interpreted as symmetrical.

[0010] [Configuration of Driving Assistance Device] FIG. 1 is a block diagram showing a driving assistance device 10 according to the present invention. The driving assistance device 10 is a group of devices that perform driving assistance for a vehicle, and, for example, drives the vehicle using autonomous driving control. As an example, the driving assistance device 10 generates a driving route to a destination set by a vehicle occupant or the like, and controls the vehicle's actuators to drive the vehicle along the driving route. Alternatively or in addition to this, the driving assistance device 10 may present information to the driver of the vehicle and assist the driver's driving operation. The driving assistance device 10 may be an in-vehicle system, or a part of it may be provided outside the vehicle.

[0011] Autonomous driving control refers to autonomously controlling the driving behavior of a vehicle using a vehicle control device, and such driving behavior includes all driving behaviors such as acceleration, deceleration, starting, stopping, and steering. Autonomously controlling driving behavior means that the control device controls driving behavior using the vehicle's devices. The control device controls these driving behaviors within a predetermined range, and driving behaviors that are not controlled by the control device are manually operated by the driver. When the vehicle is driven manually by the driver without autonomous driving control, the control device does not autonomously control the driving behavior, and the vehicle's driving behavior is controlled by the driver's operation.

[0012] As shown in Fig. 1, the driving assistance device 10 includes a map database 11, a navigation device 12, an imaging device 13, a distance measuring device 14, a display device 15, and a control device 16. The control device 16 in this embodiment also includes a trajectory generation device 20 as a part thereof. The devices constituting the driving assistance device 10 are connected via a Controller Area Network (CAN) or other in-vehicle LAN, and can exchange information with each other. Information is exchanged with devices provided outside the vehicle via a network such as the Internet or a local area network (LAN).

[0013] The map database 11 is a storage medium storing map information and is installed inside or outside the vehicle. The control device 16 acquires map information from the map database 11 as needed. The map information is used for generating a driving route, etc., and includes information on nodes corresponding to specific points on roads where the vehicle's direction of travel changes (e.g., intersections, forks, etc.) and information on links corresponding to road sections connecting the nodes. Node information includes location information (e.g., latitude and longitude) and information on entering and exiting intersections and forks. Link information includes road width, road curvature radius, road shoulder structures, road traffic regulations (e.g., traffic direction), merging points, forks, etc. Link information may include information on link lines along the road that indicate the shape of the road. The map information may also be high-precision map information (HD map) that allows the movement trajectory of each lane to be identified.

[0014] The navigation device 12 is a device that references map information and generates a driving route from the current position of the vehicle detected by a positioning system (not shown) to a set destination. The navigation device 12 uses node and link information in the map information to search for a driving route for the vehicle to reach the destination. The driving route includes at least information on the roads on which the vehicle will travel and the direction of travel of the vehicle, and is displayed, for example, using nodes and link lines. The control device 16 may display the generated driving route on the display device 15.

[0015] The imaging device 13 is a device that captures images of objects around the vehicle, and examples thereof include a camera equipped with an imaging element such as a CCD, an ultrasonic camera, an infrared camera, etc. In order to reduce blind spots when recognizing objects, multiple imaging devices 13 are installed on the front grille of the vehicle, below the left and right door mirrors, near the rear bumper, etc.

[0016] The distance measuring device 14 is a device that acquires the relative distance and relative speed between the distance measuring device 14 and an object, and examples thereof include laser radar, millimeter wave radar, and LiDAR (Light Detection and Ranging) units. In order to reduce blind spots when recognizing an object, multiple distance measuring devices 14 are installed at the front, right side, left side, and rear of the vehicle.

[0017] The target object is an object existing on or around the road, including lane boundaries, center lines, road markings, medians, guardrails, curbs, road signs, traffic lights, and crosswalks. The target object also includes obstacles that may affect the vehicle's travel, such as automobiles other than the vehicle itself, motorcycles, bicycles, and pedestrians. The control device 16 acquires the detection results of the imaging device 13 and the distance measuring device 14 at predetermined time intervals (e.g., every 0.1 to 1 millisecond). The detection results of the imaging device 13 and the distance measuring device 14 may also be integrated or synthesized (sensor fusion) by the control device 16. This allows missing information about the target object to be supplemented.

[0018] The ranging device 14 may generate point cloud data in which information on ranging points of objects present around the vehicle is arranged two-dimensionally in the left-right and up-down directions of the vehicle. The ranging points of the object are points on the object whose distances to the ranging device 14 are measured. The information on the ranging points includes not only position information of the ranging points but also the reflectivity of electromagnetic waves at the ranging points. The position information on the ranging points includes information on the coordinates of the ranging points and information on the distance from the ranging device 14 to the ranging points.

[0019] The distance measuring device 14 generates point cloud data by scanning electromagnetic waves in the left-right direction of the vehicle. For example, the distance measuring device 14 emits electromagnetic waves along the vehicle width direction, detects the reflected electromagnetic waves (reflected waves), and calculates the distance to the distance measurement point and the direction of the distance measurement point. The irradiation range of the electromagnetic waves is not particularly limited. Examples of electromagnetic waves include millimeter waves, infrared rays, and lasers.

[0020] The display device 15 is a device for providing necessary information to the vehicle occupants, and is a liquid crystal display provided on the instrument panel, a projector such as a head-up display (HUD), etc. The display device 15 may also include an input device (e.g., a touch panel) for the occupant to input instructions to the control device 16. The display device 15 may also include a speaker as an output device.

[0021] The control device 16 is a device that performs driving assistance by controlling and cooperating with the devices that make up the driving assistance device 10. The control device 16 is, for example, a computer, and includes a CPU (Central Processing Unit) that is a processor, a ROM (Read Only Memory) that stores programs, and a RAM (Random Access Memory) that functions as an accessible storage device. The CPU is an operating circuit that executes the programs stored in the ROM and realizes the functions of the control device 16.

[0022] The control device 16 has a driving assistance function that executes driving assistance for the vehicle. The control device 16 also includes a trajectory generation device 20 as a part thereof, and the trajectory generation device 20 has a trajectory generation function that generates a traveling trajectory of the vehicle as part of the driving assistance function. The ROM stores a program for realizing the driving assistance function, and the CPU executes the program stored in the ROM to realize the driving assistance function including the trajectory generation function.

[0023] 1 illustrates, for convenience, the driving control unit 17, the lane detection unit 21, the accuracy calculation unit 22, the lane selection unit 23, and the trajectory generation unit 24 as functional blocks that realize the driving assistance function. Of these functional blocks, the lane detection unit 21, the accuracy calculation unit 22, the lane selection unit 23, and the trajectory generation unit 24 are functional blocks that realize the trajectory generation function, and are included in the trajectory generation device 20. The functions of each functional block will be described below with reference to FIG. 2.

[0024] [Function of Trajectory Generation Device] Fig. 2 is a plan view showing an example of a driving scene in which the control device 16 executes autonomous driving control using the driving assistance function. In the driving scene shown in Fig. 2, two two-lane roads A1 and A2 extend in the vertical direction of the drawing, and a three-lane road A3 extends in the horizontal direction of the drawing. The intersection of roads A1 and A3 is intersection B1, and the intersection of roads A2 and A3 is intersection B2. In addition, a crosswalk C that crosses road A3 is provided on the side of intersection B1 between intersections B1 and B2.

[0025] In the driving scene shown in FIG. 2 , vehicle V is traveling at position P1 in lane La of road A1 and traveling to destination D ahead in lane Lb of road A2. Furthermore, vehicle V traveling in lane L1 of road A3 can turn left or go straight at intersection B2, vehicle V traveling in lane L2 can go straight at intersection B2, and vehicle V traveling in lane L3 can turn right or go straight at intersection B2. In this case, the control device 16, using the function of the driving control unit 17, generates a driving route that involves entering intersection B1 from position P1, turning right at intersection B1 to enter one of lanes L1 to L3 of road A3, entering intersection B2 from lane L1 and turning left, and traveling toward destination D ahead in lane Lb of road A2. Then, the control device 16 autonomously controls the driving behavior of vehicle V so that vehicle V travels along the driving route.

[0026] If the map information stored in the map database 11 is high-precision map information, lanes can be recognized from lane information for roads A1 to A3. However, if the map information does not include lane information for roads A1 to A3, the lanes for roads A1 to A3 are detected from images acquired by the imaging device 13, and a travel trajectory is generated. This lane detection that does not rely on map information may not always accurately detect lanes, which may prevent autonomous driving control from continuing. Therefore, the control device 16 of this embodiment selects a lane for the vehicle V to travel in based on the probability that the detected lane can be traveled in autonomous driving control.

[0027] The driving control unit 17 has a function of generating a driving route for the vehicle V and activating an actuator of the vehicle V so that the vehicle V travels along the driving route. The control device 16 uses the function of the driving control unit 17 to determine whether the vehicle V traveling along the driving route will enter an intersection, and if it determines that the vehicle V will enter the intersection, outputs an execution instruction to the trajectory generation device 20 to generate a driving trajectory in which the vehicle V enters an exit lane of the intersection.

[0028] When the vehicle V enters an intersection, the lane detection unit 21 has a function of detecting candidate lanes that are candidates for lanes along which the vehicle V will travel when exiting the intersection. Using the function of the lane detection unit 21, the trajectory generation device 20 obtains the traveling direction of the vehicle V when exiting the intersection from the traveling route and the current position of the vehicle V obtained from a positioning system (not shown), and detects the lane along which the vehicle V will travel after exiting the intersection as a candidate lane.

[0029] Specifically, when the vehicle V goes straight through the intersection, the traveling direction of the vehicle V after exiting the intersection is forward, and the trajectory generating device 20 detects the lane ahead in the traveling direction (beyond the intersection) relative to the intersection as a candidate lane. When the vehicle V turns right at the intersection, the traveling direction of the vehicle V after exiting the intersection is right, and the trajectory generating device 20 detects the lane on the right side of the traveling direction relative to the intersection as a candidate lane. When the vehicle V turns left at the intersection, the traveling direction of the vehicle V after exiting the intersection is left, and the trajectory generating device 20 detects the lane on the left side of the traveling direction relative to the intersection as a candidate lane. Note that the reason for using lane candidates is that the lane in which the vehicle V will travel when exiting the intersection has not been determined at the time of lane detection.

[0030] The trajectory generating device 20 detects candidate lanes from the detection results of a detection device that detects lanes. The detection device that detects lanes is, for example, at least one of an imaging device 13 mounted on the vehicle V that captures images of the surroundings of the vehicle V, and a ranging device 14 mounted on the vehicle V that generates point cloud data. The trajectory generating device 20 detects candidate lanes from images acquired from the imaging device 13, and may alternatively or additionally detect candidate lanes from point cloud data acquired from the ranging device 14.

[0031] Specifically, the trajectory generating device 20 performs processes such as edge extraction, semantic segmentation, and pattern matching on the image acquired from the imaging device 13, and extracts from the image portions corresponding to lanes separated by lane boundaries, etc. The trajectory generating device 20 may also acquire position information of ranging points from the point cloud data acquired from the ranging device 14, generate an image on which the ranging points are plotted, and perform processes such as pattern matching on the image to extract portions corresponding to lanes.

[0032] For example, the trajectory generating device 20 recognizes lanes by detecting structures that define lanes, such as curbs, medians, and guardrails, from an image on which distance measurement points are plotted. The trajectory generating device 20 may also detect lane boundaries from information on the reflectance of electromagnetic waves, utilizing the difference in reflectance between the road surface and lane boundary (white line) portions. Note that lane boundaries refer to boundaries marked on the road surface, and are different from structures that define lanes, such as medians.

[0033] 2, a travel route is set in which vehicle V turns right at intersection B1, so trajectory generating device 20 detects the lane on the right side of intersection B1 in the traveling direction of vehicle V as a candidate lane. Assuming that curb X1, white lines X2 and X3, and center divider X4 are detected from the detection results of imaging device 13 and distance measuring device 14, trajectory generating device 20 detects lane L1 defined by curb X1 and white line X2, lane L2 defined by white lines X2 and X3, and lane L3 defined by white line X3 and center divider X4 as candidate lanes.

[0034] The accuracy calculation unit 22 has a function of calculating the accuracy that the vehicle V can travel along the candidate lane. The accuracy that the vehicle V can travel along the candidate lane is, in particular, the accuracy that the vehicle V can travel along the candidate lane under autonomous driving control, and more specifically, the accuracy that the vehicle V exiting an intersection can enter the candidate lane under autonomous driving control and travel along the candidate lane. In other words, this accuracy is the accuracy that the vehicle V can pass through the intersection under autonomous driving control and enter the candidate lane, and the accuracy that the vehicle V that entered the intersection under autonomous driving control can exit the intersection and enter the candidate lane while continuing autonomous driving control.

[0035] The probability that a candidate lane can be traveled through by autonomous driving control is calculated based on the probability that a lane actually exists at the position where the candidate lane is detected, the probability that a lane boundary line defining the candidate lane can be detected, the probability that a structure actually exists at the position where a structure defining the candidate lane is detected, and the probability that the direction of travel of the candidate lane is aligned with the direction of travel of the vehicle V. For example, the probability of a candidate lane where there is no boundary line between the sidewalk and the roadway is calculated to be lower than the probability of a candidate lane where there is a boundary line between the sidewalk and the roadway, and the probability of a candidate lane where the lane boundary line is interrupted midway is calculated to be lower than the probability of a candidate lane where the lane boundary line is not interrupted.

[0036] As an example, the trajectory generating device 20 calculates the probability of a candidate lane for which lane boundary lines on both sides are detected to be higher than the probability of a candidate lane for which lane boundary lines on at least one side are not detected, using the function of the probability calculation unit 22. For example, in the driving scene shown in Figure 2, the probability of lane L2 when white lines X2 and X3 can be accurately detected is calculated to be higher than the probability of lane L2 when white line X2 has disappeared due to wear and cannot be detected.

[0037] The trajectory generation device 20 may calculate the accuracy of a candidate lane that is separated by a structure on at least one side to be higher than the accuracy of a candidate lane that is separated by lane boundary lines on both sides. This is because a three-dimensional object such as a structure is less likely to be erroneously detected than a lane boundary line such as a white line. For example, in the driving scene shown in Figure 2, the accuracy of lane L3, which is separated by a center divider X4 on the right side, is calculated to be higher than the accuracy of lane L2, which is separated by white lines X2 and X3 on both sides.

[0038] Furthermore, if a crosswalk is present near the exit where vehicle V exits the intersection, the trajectory generation device 20 may calculate the likelihood of a candidate lane closer to the edge of the crosswalk in the road width direction to be higher than the likelihood of a candidate lane farther from the edge of the crosswalk. This is because the closer to the center of the crosswalk, the more difficult it is to distinguish between the own lane and the oncoming lane. For example, in the driving scene shown in FIG. 2 , a crosswalk C is present near the exit E when vehicle V passes through intersection B1 along the driving route. In this case, the likelihood of lane L1 close to the edge on the left side of crosswalk C in the traveling direction is calculated to be higher than the likelihood of lanes L2 and L3 farther from the edge on the left side of crosswalk C in the traveling direction (i.e., closer to the center of crosswalk C).

[0039] Furthermore, when there is a preceding vehicle traveling ahead of vehicle V, trajectory generation device 20 may calculate the probability of a candidate lane in which the preceding vehicle has traveled higher than the probability of a candidate lane in which the preceding vehicle has not traveled. This is because there is a high probability that vehicle V can travel in the lane in which the preceding vehicle has traveled. For example, in the traveling scene shown in FIG. 2 , when there is a preceding vehicle Vx traveling at position Px of lane L2, the probability of lane L2 is calculated to be higher than the probabilities of lanes L1 and L3.

[0040] In particular, when there are multiple candidate lanes, the trajectory generating device 20 calculates, for each candidate lane, the probability that the vehicle V can travel through the candidate lane by autonomous driving control. On the other hand, when there is only one candidate lane, the trajectory generating device 20 may or may not calculate the probability. Note that when there are multiple candidate lanes, the trajectory generating device 20 may calculate the probability for only some of the candidate lanes.

[0041] 3 is an example of a flowchart showing information processing executed in the trajectory generating device 20 when the above-mentioned accuracy is calculated by the function of the accuracy calculation unit 22. The processing described below is executed by the processor (CPU) of the control device 16.

[0042] First, in step S1, the accuracy of the candidate lane for which accuracy is calculated is set to "0", and then in step S2, it is determined whether or not the boundary on both sides of the candidate lane is defined. If it is determined that the boundary on both sides of the candidate lane is defined, "8" is added to the accuracy of the candidate lane. If it is determined that the boundary on at least one side of the candidate lane is not defined, the process proceeds to step S4.

[0043] In step S4, it is determined whether at least one side of the candidate lane is defined by a structure. If it is determined that at least one side of the candidate lane is defined by a structure, the process proceeds to step S5, where "6" is added to the accuracy of the candidate lane. If it is determined that both sides of the candidate lane are not defined by a structure, the process proceeds to step S6.

[0044] In step S6, it is determined whether the candidate lane is closer to the edge of the crosswalk in the road width direction. If it is determined that the candidate lane is closer to the edge of the crosswalk in the road width direction, the process proceeds to step S7, where "4" is added to the accuracy of the candidate lane. If it is determined that the candidate lane is farther from the edge of the crosswalk in the road width direction, the process proceeds to step S8.

[0045] In step S8, it is determined whether or not a preceding vehicle Vx is present in the candidate lane. If it is determined that a preceding vehicle Vx is present in the candidate lane, the process proceeds to step S9, where "10" is added to the accuracy of the candidate lane. If it is determined that a preceding vehicle Vx is not present in the candidate lane, the process proceeds to step S10.

[0046] In step S10, it is determined whether the accuracy has been calculated for all candidate lanes. If it is determined that the accuracy has been calculated for all candidate lanes, the process ends. If it is determined that the accuracy has not been calculated for some candidate lanes, the process proceeds to step S1.

[0047] When the accuracy of lanes L1 to L3 shown in FIG. 2 is calculated according to the flowchart shown in FIG. 3, the accuracy of lane L1 is increased by "8" due to the curb X1 and white line X2 (step S3), "6" due to the curb X1 (step S5), and "4" due to lane L2 being closer to the edge of the crosswalk C in the road width direction than lanes L2 and L3 (step S7), for a total of "18." The accuracy of lane L2 is increased by "8" due to the white lines X2 and X3 (step S3), and "10" due to the presence of leading vehicle Vx (step S9), for a total of "18." The accuracy of lane L3 is increased by "8" due to the white line X3 and center divider X4 (step S3), and "6" due to the center divider X4 (step S5), for a total of "14."

[0048] The lane selection unit 23 has a function of selecting, based on the calculated probability, from the candidate lanes, an exit lane in which the vehicle V will actually travel when exiting the intersection. For example, the trajectory generation device 20 uses the function of the lane selection unit 23 to select, as the exit lane, the candidate lane that has the highest probability that the vehicle V will be able to travel in the candidate lane under autonomous driving control.

[0049] 2, the lanes L1 and L2 have the same accuracy value, but the trajectory generating device 20 recognizes that the lane L2 where the preceding vehicle Vx is located has a higher accuracy than the lane L1, and selects the lane L2 as the exit lane. Note that if there is only one candidate lane, the trajectory generating device 20 sets the only candidate lane as the exit lane along which the vehicle V will travel when exiting the intersection.

[0050] The trajectory generating device 20 may also select an exit lane based on the distance between a first intersection located ahead of the vehicle V and a second intersection that the vehicle V will enter after the first intersection. For example, the trajectory generating device 20 calculates the distance between a first intersection located ahead of the vehicle V and that the vehicle V will enter, and a second intersection that the vehicle V will enter after the first intersection. If the distance between the first intersection and the second intersection is equal to or greater than a predetermined value, the trajectory generating device 20 selects the candidate lane with the highest probability as the exit lane.

[0051] On the other hand, if the distance between the first intersection and the second intersection is less than a predetermined value, the candidate lane that minimizes the number of lane changes required by the vehicle V between the first intersection and the second intersection is selected as the exit lane. Alternatively or in addition, if the distance between the first intersection and the second intersection is less than a predetermined value, the candidate lane that minimizes the number of lane changes required by the vehicle V between the first intersection and the second intersection may be selected as the exit lane from among candidate lanes whose probability of being able to travel on the candidate lane using autonomous driving control is equal to or greater than a predetermined threshold. However, when traveling straight through the second intersection, the candidate lane with the highest probability may be selected as the exit lane even if the distance between the first intersection and the second intersection is less than the predetermined value. The predetermined threshold can be set to an appropriate value within a range in which the vehicle V can continue traveling using autonomous driving control.

[0052] The predetermined value can be set appropriately based on the travel distance (e.g., 15 to 30 m) required for vehicle V to change lanes. The first intersection and the second intersection are recognized from the position information of the nodes on the travel route and the current position of vehicle V obtained from a positioning system (not shown). The distance between the first intersection and the second intersection is obtained from the road information of the link between the nodes corresponding to the first intersection and the second intersection.

[0053] 2, if the distance between intersection B1, which is the first intersection, and intersection B2, which is the second intersection, is equal to or greater than a predetermined value, the trajectory generating device 20 selects lane L2, which has the highest probability of being able to travel through the candidate lanes under autonomous driving control, as the exit lane. On the other hand, if the distance between intersections B1 and B2 is less than the predetermined value, the trajectory generating device 20 selects lane L1 as the exit lane to avoid a lane change and turn left at intersection B2.

[0054] The trajectory generation unit 24 has a function of generating a driving trajectory of the vehicle V entering the selected exit lane. The trajectory generation device 20 uses the function of the trajectory generation unit 24 to generate a driving trajectory of the vehicle V traveling from the current position to the exit lane based on the detection results of objects around the vehicle V, the minimum turning radius of the vehicle V, etc. The generated driving trajectory is output to the driving control unit 17. In addition to this, the trajectory generation device 20 may display the generated driving trajectory on the display device 15.

[0055] 4 is a plan view showing the travel trajectory generated by the trajectory generation device 20 when the distance between intersections B1 and B2 is equal to or greater than a predetermined value in the travel scene shown in FIG. 2. The trajectory generation device 20 generates travel trajectories T1 and T2 that lead from a current position P1 to a lane L2. The control device 16, using the function of the travel control unit 17, causes the vehicle V to travel from position P1 on lane La to position P2 on lane La along the travel trajectory T1, and causes the vehicle V to travel from position P2 on lane La to position P3 on lane L2 along the travel trajectory T2.

[0056] The control device 16 then generates a driving trajectory T3 in which the vehicle V changes lanes from lane L2 to lane L1. The control device 16 causes the vehicle V to travel along the driving trajectory T3 from position P2 on lane L2 to position P4 on lane L1. The control device 16 then generates a driving trajectory T4 in which the vehicle V turns left at intersection B2 and enters lane Lb. The control device 16 causes the vehicle V to travel along the driving trajectory T4 from position P4 on lane L1 to position P5 on lane Lb. After the vehicle V enters lane Lb, the control device 16 controls the driving operation of the vehicle V so that the vehicle V autonomously travels along lane Lb to destination D.

[0057] In contrast, Fig. 5 is a plan view showing the travel trajectory generated by the trajectory generation device 20 when the distance between intersections B1 and B2 is less than a predetermined value in the travel scene shown in Fig. 2. The trajectory generation device 20 generates travel trajectories T1 and T5 that enter lane L1 from position P1, which is the current position. The control device 16, using the function of the travel control unit 17, causes the vehicle V to travel along the travel trajectory T1 from position P1 on lane La to position P2, and also causes the vehicle V to travel along the travel trajectory T5 from position P2 on lane La to position P6 on lane L1.

[0058] The control device 16 then generates a driving trajectory T6 in which the vehicle V drives along the lane L1. The control device 16 causes the vehicle V to drive along the driving trajectory T6 from position P6 on the lane L1 to position P4. The control device 16 then generates a driving trajectory T4 in which the vehicle V turns left at intersection B2 and enters lane Lb. The control device 16 causes the vehicle V to drive along the driving trajectory T4 from position P4 on the lane L1 to position P5 on the lane Lb. After the vehicle V enters lane Lb, the control device 16 controls the driving operation of the vehicle V so that the vehicle V drives autonomously along lane Lb to destination D.

[0059] 6 and 7A-7B, the procedure for processing information by the control device 16 will be described. The process described below is executed by a processor (CPU) included in the control device 16 at predetermined time intervals (for example, every 0.1 to 1 millisecond).

[0060] FIG. 6 is an example of a flowchart showing information processing executed by the driving assistance device 10 of this embodiment.

[0061] First, in step S11, the lane detection unit 21 detects candidate lanes, and in the following step S12, the accuracy calculation unit 22 determines whether there are multiple candidate lanes. If it is determined that there are multiple candidate lanes, the process proceeds to step S13, where the accuracy of traveling through the candidate lanes by autonomous driving control is calculated. In the following step S14, the lane selection unit 23 selects an exit lane from the candidate lanes based on the accuracy. On the other hand, if it is determined that there is only one candidate lane, the process proceeds to step S15, where the candidate lane is set as the exit lane. Then, in step S16, a driving trajectory for entering the selected exit lane is generated.

[0062] Next, FIGS. 7A and 7B are another example of a flowchart showing information processing executed in the driving assistance device 10 of this embodiment.

[0063] 7A, the driving control unit 17 generates a driving route from the current position to the destination D, and drives the vehicle V along the driving route. In the subsequent step S22, it is determined whether the vehicle V will enter an intersection. If it is determined that the vehicle V will not enter an intersection, the process proceeds to step S21. On the other hand, if it is determined that the vehicle V will enter an intersection, the process proceeds to step S23.

[0064] In step S23, the lane detection unit 21 detects candidate lanes, and in the following step S24, the accuracy calculation unit 22 determines whether there are multiple candidate lanes. If it is determined that there are multiple candidate lanes, the process proceeds to step S25, where the accuracy of traveling through the candidate lane by autonomous driving control is calculated for each candidate lane, and the process proceeds to step S27 in Fig. 7B. On the other hand, if it is determined that there is only one candidate lane, the process proceeds to step S26, where the candidate lane is set as an exit lane, and the process proceeds to step S30 in Fig. 7B.

[0065] 7B, the lane selector 23 calculates the distance between the first intersection and the second intersection, and then determines whether the distance between the intersections is less than a predetermined value in step S28. If it is determined that the distance between the intersections is equal to or greater than the predetermined value, the process proceeds to step S29, where the candidate lane with the highest probability is selected as the exit lane.

[0066] On the other hand, if it is determined that the distance between the intersections is less than the predetermined value, the process proceeds to step S30, where it is determined whether the vehicle V will go straight through the second intersection. If it is determined that the vehicle V will not go straight through the second intersection, the process proceeds to step S29. On the other hand, if it is determined that the vehicle V will go straight through the second intersection, the process proceeds to step S31, where the candidate lane that minimizes the number of lane changes between the first intersection and the second intersection is selected as the exit lane. Then, in step S32, the trajectory generation unit 24 generates a driving trajectory for entering the selected exit lane.

[0067] Note that step S15 in Fig. 6 is not essential to the present invention and may be omitted as necessary. Also, steps S21, S22, and S26 in Fig. 7A and steps S27, S28, S30, and S31 in Fig. 7B are not essential to the present invention and may be omitted as necessary.

[0068] According to this embodiment, in a trajectory generation method executed by a trajectory generation device, when a vehicle V enters an intersection, the trajectory generation device 20 detects candidate lanes that are candidates for lanes along which the vehicle V will travel when exiting the intersection from the detection results of a lane detection device, and when there are multiple candidate lanes, calculates a probability that the vehicle V will be able to travel in the candidate lanes through autonomous driving control, and based on the probability, selects from the candidate lanes an exit lane along which the vehicle V will travel when exiting the intersection, and generates a driving trajectory of the vehicle V entering the exit lane. This allows the vehicle V to continue autonomous driving along the lane after exiting the intersection.

[0069] In the trajectory generation method of this embodiment, the trajectory generation device 20 may calculate the probability of the candidate lane for which lane boundary lines on both sides are detected to be higher than the probability of the candidate lane for which no lane boundary line is detected on at least one side, thereby allowing the vehicle V to enter a lane in which it is easier to continue autonomous driving.

[0070] In the trajectory generation method of this embodiment, the trajectory generation device 20 may calculate the accuracy of the candidate lane that is separated on at least one side by a structure to be higher than the accuracy of the candidate lane that is separated on both sides by lane boundary lines, thereby allowing the vehicle V to enter a lane that is more likely to continue autonomous driving.

[0071] In the trajectory generation method of this embodiment, if a pedestrian crossing is present near the exit where the vehicle V exits the intersection, the trajectory generation device 20 may calculate the accuracy of the candidate lane closer to the edge of the pedestrian crossing in the road width direction to be higher than the accuracy of the candidate lane farther from the edge of the pedestrian crossing. This allows the vehicle V to enter a lane where it is easier to continue autonomous driving.

[0072] In the trajectory generation method of this embodiment, when there is a preceding vehicle Vx traveling ahead of the vehicle V, the trajectory generation device 20 may calculate the probability of the candidate lane in which the preceding vehicle Vx has traveled to be higher than the probability of the candidate lane in which the preceding vehicle Vx has not traveled, thereby enabling the vehicle V to enter a lane in which it is easier to continue autonomous driving.

[0073] In the trajectory generation method of this embodiment, the trajectory generation device 20 calculates the distance between a first intersection B1 located ahead of the vehicle V and a second intersection B2 that the vehicle V will enter after the first intersection B1, and if the distance is equal to or greater than a predetermined value, the candidate lane with the highest probability may be selected as the exit lane. This allows the vehicle V to enter a lane in which it is easier to continue autonomous driving.

[0074] In the trajectory generation method of this embodiment, the trajectory generation device 20 calculates the distance between a first intersection B1 located ahead of the vehicle V and a second intersection B2 that the vehicle V will enter after the first intersection B1, and if the distance is less than a predetermined value, may select as the exit lane the candidate lane that minimizes the number of lane changes required by the vehicle V between the first intersection B1 and the second intersection B2. This makes it possible to prevent the vehicle V from completing lane changes between intersections and being unable to travel along the travel route.

[0075] In the trajectory generation method of this embodiment, the trajectory generation device 20 may calculate the distance between a first intersection B1 located ahead of the vehicle V and a second intersection B2 that the vehicle V will enter after the first intersection B1, and if the distance is less than a predetermined value, select, as the exit lane, from among the candidate lanes whose accuracy is equal to or greater than a predetermined threshold, the candidate lane that minimizes the number of lane changes required by the vehicle V between the first intersection B1 and the second intersection B2. This makes it possible to prevent the occurrence of a situation in which lane changes are not completed between intersections, making it impossible to travel along the travel route.

[0076] Furthermore, according to this embodiment, a trajectory generation device 20 is provided, which includes: a lane detection unit 21 that, when a vehicle V enters an intersection, detects candidate lanes that are candidates for lanes along which the vehicle V will travel when exiting the intersection, based on the detection results of a detection device that detects lanes; a probability calculation unit 22 that, when there are multiple candidate lanes detected by the lane detection unit 21, calculates a probability that the vehicle V will be able to travel on the candidate lanes by autonomous driving control; a lane selection unit 23 that selects, from the candidate lanes based on the probability calculated by the probability calculation unit 22, an exit lane along which the vehicle V will travel when exiting the intersection; and a trajectory generation unit 24 that generates a travel trajectory of the vehicle V entering the exit lane selected by the lane selection unit 23. This allows the vehicle V to continue autonomous driving along the lanes after exiting the intersection.

[0077] DESCRIPTION OF SYMBOLS 10... Driving assistance device 11... Map database 12... Navigation device 13... Imaging device 14... Distance measuring device 15... Display device 16... Control device 17... Driving control unit 20... Trajectory generation device 21... Lane detection unit 22... Accuracy calculation unit 23... Lane selection unit 24... Trajectory generation unit A1, A2, A3... Road B1, B2... Intersection C... Pedestrian crossing D... Destination E... Exit L1, L2, L3, La, Lb... Lanes P1, P2, P3, P4, P5, P6, Px... Position T1, T2, T3, T4, T5, T6... Driving trajectory V... Vehicle Vx... Leading vehicle X1... Curb X2, X3... White line X4... Median strip

Claims

1. In a trajectory generation method performed by a trajectory generation device, The trajectory generation device is When a vehicle enters an intersection, the detection results of the lane detection device detect candidate lanes, which are candidates for the lane the vehicle will travel in when exiting the intersection. If there are multiple candidate lanes, the probability that the vehicle can travel in one of the candidate lanes using autonomous driving control is calculated. Based on the aforementioned accuracy, select the exit lane from the candidate lanes that the vehicle will use when exiting the intersection. The system generates a driving trajectory of the vehicle entering the exit lane. A trajectory generation method that calculates the accuracy of a candidate lane that is separated on at least one side by a structure to be higher than the accuracy of a candidate lane that is separated on both sides by lane boundary lines.

2. In a trajectory generation method performed by a trajectory generation device, The trajectory generation device is When a vehicle enters an intersection, the detection results of the lane detection device detect candidate lanes, which are candidates for the lane the vehicle will travel in when exiting the intersection. If there are multiple candidate lanes, the probability that the vehicle can travel in one of the candidate lanes using autonomous driving control is calculated. Based on the aforementioned accuracy, select the exit lane from the candidate lanes that the vehicle will use when exiting the intersection. The system generates a driving trajectory of the vehicle entering the exit lane. A trajectory generation method in which, if a pedestrian crossing exists near the exit where the vehicle exits the intersection, the accuracy of the candidate lane closer to the road width direction of the pedestrian crossing is calculated to be higher than the accuracy of the candidate lane further from the edge of the pedestrian crossing.

3. In a trajectory generation method performed by a trajectory generation device, The trajectory generation device is When a vehicle enters an intersection, the detection results of the lane detection device detect candidate lanes, which are candidates for the lane the vehicle will travel in when exiting the intersection. If there are multiple candidate lanes, the probability that the vehicle can travel in one of the candidate lanes using autonomous driving control is calculated. Based on the aforementioned accuracy, select the exit lane from the candidate lanes that the vehicle will use when exiting the intersection. The system generates a driving trajectory of the vehicle entering the exit lane. The distance between the first intersection located in front of the vehicle and the second intersection that the vehicle enters after the first intersection is calculated. A trajectory generation method that, if the distance is greater than or equal to a predetermined value, selects the candidate lane with the highest accuracy as the exit lane.

4. The trajectory generation method according to any one of claims 1 to 3, wherein the trajectory generation device calculates the accuracy of the candidate lane for which lane boundary lines on both sides are detected to be higher than the accuracy of the candidate lane for which at least one lane boundary line is not detected.

5. The trajectory generation method according to any one of claims 1 to 4, wherein, if there is a preceding vehicle traveling in front of the vehicle, the trajectory generation device calculates the accuracy of the candidate lane traveled by the preceding vehicle to be higher than the accuracy of the candidate lane not traveled by the preceding vehicle.

6. The trajectory generation device is The distance between the first intersection located in front of the vehicle and the second intersection that the vehicle enters after the first intersection is calculated. The trajectory generation method according to any one of claims 1 to 5, wherein if the distance is less than a predetermined value, the candidate lane that results in the fewest number of lane changes by the vehicle between the first intersection and the second intersection is selected as the exit lane.

7. The trajectory generation device is The distance between the first intersection located in front of the vehicle and the second intersection that the vehicle enters after the first intersection is calculated. If the distance is less than a predetermined value, the trajectory generation method according to any one of claims 1 to 6, wherein the candidate lane with an accuracy of equal to or greater than a predetermined threshold is selected as the exit lane, which is the candidate lane that results in the fewest number of lane changes by the vehicle between the first intersection and the second intersection.

8. A lane detection unit detects candidate lanes, which are candidates for the lane the vehicle will travel in when it exits the intersection, based on the detection results of a lane detection device when the vehicle enters an intersection. When the lane detection unit detects multiple candidate lanes, the probability calculation unit calculates the probability that the vehicle can travel in the candidate lane using autonomous driving control. A lane selection unit selects an exit lane from the candidate lanes based on the accuracy calculated by the accuracy calculation unit, which the vehicle will use when exiting the intersection. The system includes a trajectory generation unit that generates a driving trajectory in which the vehicle enters the exit lane selected by the lane selection unit, The accuracy calculation unit calculates the accuracy of the candidate lane, which is separated by a structure on at least one side, to be higher than the accuracy of the candidate lane, which is separated by lane boundary lines on both sides.

9. A lane detection unit detects candidate lanes, which are candidates for the lane the vehicle will travel in when it exits the intersection, based on the detection results of a lane detection device when the vehicle enters an intersection. When the lane detection unit detects multiple candidate lanes, the probability calculation unit calculates the probability that the vehicle can travel in the candidate lane using autonomous driving control. A lane selection unit selects an exit lane from the candidate lanes based on the accuracy calculated by the accuracy calculation unit, which the vehicle will use when exiting the intersection. The system includes a trajectory generation unit that generates a driving trajectory in which the vehicle enters the exit lane selected by the lane selection unit, The accuracy calculation unit calculates the accuracy of the candidate lane closer to the road width direction end of the pedestrian crossing when a pedestrian crossing exists near the exit where the vehicle exits the intersection to be higher than the accuracy of the candidate lane further from the end of the pedestrian crossing.

10. A lane detection unit detects candidate lanes, which are candidates for the lane the vehicle will travel in when it exits the intersection, based on the detection results of a lane detection device when the vehicle enters an intersection. When the lane detection unit detects multiple candidate lanes, the probability calculation unit calculates the probability that the vehicle can travel in the candidate lane using autonomous driving control. A lane selection unit selects an exit lane from the candidate lanes based on the accuracy calculated by the accuracy calculation unit, which the vehicle will use when exiting the intersection. The system includes a trajectory generation unit that generates a driving trajectory in which the vehicle enters the exit lane selected by the lane selection unit, The lane selection unit calculates the distance between the first intersection located in front of the vehicle and the second intersection that the vehicle will enter after the first intersection. A trajectory generating device that, if the distance is greater than or equal to a predetermined value, selects the candidate lane with the highest accuracy as the exit lane.