Driving control method and driving control device

Abstract

A driving control method that is used in a processor (10), wherein when it is determined, on the basis of ego vehicle information and other vehicle information which were acquired using a sensor (2), that a condition that the distance between an ego vehicle (V1) and a following vehicle (V2) traveling behind the ego vehicle (V1) is less than a prescribed distance and the acceleration of the following vehicle (V2) is greater than the acceleration of the ego vehicle (V1) is not fulfilled, a first trajectory (TJ21) is calculated in which the distance along the direction of progression in a predicted trajectory to the completion of a lane change by the following vehicle (V2) is a first distance (L1), and when it is determined that said condition is fulfilled, a second trajectory (TJ22) is calculated in which the distance along the direction of progression in a predicted trajectory (TJ) to the completion of a lane change by the following vehicle (V2) is a second distance (L2) that is shorter than the first distance (L1), and a target trajectory (TJ1) is calculated in which a movement region of the ego vehicle (V1) traveling on the target trajectory (TJ1) does not interfere with a movement region of the following vehicle (V2) traveling on the first trajectory (TJ21) or the second trajectory (TJ22).

Description

Operation control method and operation control device

[0001] The present invention relates to an autonomous operation control method and an operation control device.

[0002] 2. Description of the Related Art A driving control is known that causes a host vehicle to change lanes to a recommended lane when it is determined that a following vehicle is rapidly approaching the host vehicle based on the inter-vehicle distance between the host vehicle and the following vehicle.

[0003] International Publication No. 2018 / 123344

[0004] However, if the following vehicle also changes lanes at the same time as the host vehicle changes lanes, the proximity state between the host vehicle and the following vehicle may not be resolved.

[0005] The problem to be solved by the present invention is to resolve the situation where the host vehicle and the following vehicle are close to each other.

[0006] The present invention solves the above problem by calculating a predicted trajectory of the following vehicle that has a shorter distance along the direction of travel until the lane change is completed than when the distance between the vehicle and the following vehicle is less than a predetermined distance and the acceleration of the following vehicle is higher than the acceleration of the vehicle.

[0007] According to the present invention, it is possible to eliminate the close proximity state between the host vehicle and the following vehicle.

[0008] FIG. 1 is a block diagram showing a hardware configuration of a driving control system. FIG. 2 is a flowchart showing a driving control process. FIG. 3 is a diagram for explaining a process for calculating a predicted trajectory of a following vehicle. FIG. 4 is a diagram for explaining a process for calculating a target trajectory of the vehicle. FIG. 5 is a first diagram for explaining a process for predicting the extension direction of a predicted trajectory. FIG. 6 is a second diagram for explaining a process for predicting the extension direction of a predicted trajectory. FIG. 7 is a third diagram for explaining a process for predicting the extension direction of a predicted trajectory. FIG. 8 is a diagram showing an example of a lane change condition.

[0009] 1 shows the configuration of a driving control system 100 equipped with a vehicle driving control device 1 according to this embodiment. The driving control method is implemented using the devices (hardware) of the perception system, judgment and control system, and information system of the driving control system 100, which includes a processor 10 of the driving control device 1.

[0010] As perceptual system devices, the driving control system 100 has one or more sensor(s) 2, a host vehicle information acquisition device 3, and an other vehicle information acquisition device 4. The processor 10 acquires observed values ​​from each of the perceptual system devices and estimates the state and changes in the position, attitude, and motion (speed, acceleration, etc.) of objects including the host vehicle and other vehicles using known methods such as a Kalman filter or multiple hypothesis tracking.

[0011] A plurality of sensors 2 are installed on the vehicle, forming a sensor group. The sensors 2 detect the presence or absence of other vehicles around the vehicle, as well as their relative speeds and relative accelerations. The sensors 2 include a Global Positioning System (GPS) unit, a gyro sensor, a vehicle speed sensor, and other devices, and detect the position of the vehicle at each timing. The sensors 2 include one or more cameras installed on the vehicle. The cameras include image sensors with imaging elements such as CCDs, ultrasonic cameras, and infrared cameras. Each camera captures images of the lane patterns around the vehicle (front, rear, and sides), lane markings on the road surface, and roadside signs, and obtains lane attributes and traffic rules from the captured images. The sensors 2 include ranging sensors that detect the presence, position, and position changes of objects around the vehicle. The ranging sensors include laser radar, millimeter-wave radar (e.g., LRF), LiDAR (light detection and ranging), and ultrasonic radar. Each sensor 2 acquires information from on-board devices and external devices according to its respective function and transmits the acquired detection information to the processor 10.

[0012] The host vehicle information acquisition device 3 calculates the current position, attitude, speed, acceleration, behavior, and direction of travel of the host vehicle based on the information acquired from the sensor 2, and provides the calculated information to the processor 10. The other vehicle information acquisition device 4 calculates the position, attitude, speed, acceleration, behavior, and direction of travel of objects, including other vehicles, around the host vehicle based on the information acquired from the sensor 2, and provides the calculated information to the processor 10. Other vehicles include following vehicles traveling behind the host vehicle, and other vehicles traveling to the side of the host vehicle or to the side of the following vehicle.

[0013] The driving control system 100 includes map information 5 and lane information 51 as information-related devices. The map information 5 and lane information 51 are stored in a storage device accessible by the processor 10 via the communication device 30 and an external server. The map information 5 is high-precision map information that includes the lane information 51 referenced when executing autonomous lane change control. The lane information 51 includes identification information for identifying each of multiple lanes belonging to a road, attribute information indicating whether the lane is an overtaking lane, and traffic rule information indicating whether the lane is a no-overtaking lane. The lane attributes can be acquired from the lane information 51 stored in a traffic system server accessible by the processor 10 via the communication device 30. The navigation device 6 references the map information 5 and calculates a route to a set destination. This route includes a target trajectory. The calculated route is provided to the vehicle controller 200 and used for autonomous driving control.

[0014] The driving control system 100 includes a driving control device 1 and a vehicle controller 200 as decision-making and control devices. The driving control device 1 controls autonomous driving, causing the vehicle to travel along a target trajectory. The processor 10 includes a ROM (Read Only Memory) 12 that stores a program for controlling autonomous driving, a CPU (Central Processing Unit) 11 that executes the program stored in the ROM 12, and a RAM (Random Access Memory) 13 that functions as an accessible storage device. The processor 10 executes each function in cooperation with the hardware shown in FIG. 1 and software for implementing autonomous driving functions, including at least a function for calculating the trajectory of a following vehicle, a function for calculating a target trajectory of the vehicle, and a function for lane change control.

[0015] The processor 10's following vehicle trajectory calculation function calculates a predicted trajectory of a following vehicle when it recognizes a following vehicle traveling behind the host vehicle based on detection information from the sensor 2. The predicted trajectory includes the trajectory of the following vehicle until it completes a lane change. The processor 10 determines whether the following vehicle is rapidly approaching the host vehicle based on predetermined conditions. The processor 10 determines whether the predetermined conditions are met and calculates two different trajectories as predicted trajectories depending on the determination result. The "predicted trajectory" includes a "first trajectory" and a "second trajectory." The two trajectories differ in the distance along the traveling direction of the predicted trajectory until the following vehicle completes a lane change. The traveling direction is the extension direction of the lane (road) on which the following vehicle is traveling on the predicted trajectory. It can also be said to be the extension direction of the lane (road) on which the host vehicle is traveling. The first trajectory is the distance until the following vehicle completes a lane change, which is the first distance, and the second trajectory is the second distance, which is shorter than the first distance. The starting point of the predicted trajectory is the position of the following vehicle at the time when it is determined whether the conditions are satisfied or not. The end point of the predicted trajectory is the point at which the following vehicle has completed the lane change, i.e., the point at which the predetermined lateral position change required for the lane change is completed. The end point may be the point at which the steering amount of the following vehicle from the starting point becomes equal to or greater than a predetermined amount, the lateral position of the following vehicle moves into the adjacent lane, the steering amount becomes equal to or less than the predetermined amount, and the direction of the following vehicle coincides with the extension direction of the adjacent lane. The end point may be the point at which the entire vehicle body enters the adjacent lane, or the point at which all wheels cross the boundary line separating the adjacent lane. The first trajectory is relatively longer than the second trajectory and is a trajectory in which the following vehicle moves laterally gradually into the adjacent lane. The lateral displacement of the first trajectory and the second trajectory is the same. The second trajectory is a trajectory in which the following vehicle closes the distance to the host vehicle and then quickly moves laterally into the adjacent lane. The second track has a larger lateral displacement (lateral displacement speed) relative to the longitudinal displacement and a larger inclination of the track during lateral displacement than the first track. The steering change amount and lateral G force of the vehicle when changing lanes on the second track are larger than the steering change amount and lateral G force of the vehicle when changing lanes on the first track.

[0016] The target trajectory calculation function of the processor 10 calculates a target trajectory of the host vehicle based on the calculated predicted trajectory of the following vehicle. The target trajectory is calculated so that the movement area of ​​the host vehicle does not interfere with the movement area of ​​the following vehicle traveling on the predicted trajectory. Because both the host vehicle and other vehicles physically occupy the road surface, when considering interference / non-interference between the host vehicle and the following vehicle, it is necessary to consider the physical occupied areas of the host vehicle and the following vehicle and a reserve area corresponding to the control error. In this embodiment, a target trajectory based on linear information is calculated as the control target value, and a target trajectory based on band-shaped area information is calculated as the limit value (tolerance value). The movement area in this embodiment is defined based on the occupied area of ​​the vehicle, including the host vehicle and the following vehicle, during movement. The occupied area is determined according to the size of the vehicle. The size of the vehicle can be determined according to the vehicle type. The vehicle type, such as a compact car, a standard car, a large car, a truck, or a motorcycle, can be determined from images captured by the camera used as the sensor 2. In addition to the occupied area, the movement area may also include a reserve area in which vehicles are allowed to approach each other. In other words, the movement area may be determined by adding the reserve area to the outer periphery of the occupied area. Interference occurs when two moving areas approach each other by less than a predetermined distance. When a moving area includes a reserve area, interference can be defined as two moving areas not overlapping each other. The target trajectory includes position information and command values ​​for autonomous driving control associated with the position or time. The command values ​​include a speed command value for driving the vehicle. The processor 10 calculates optimal command values ​​taking into account the distance to the preceding vehicle and the relative speed. The processor 10 calculates the target position, attitude, speed, and curvature of the vehicle dt seconds from the set speed and target trajectory. The target trajectory is generated by calculating this for N steps from the current time t to a future time t + N*dt. In this way, the target trajectory includes the positions the vehicle will follow and the times it will pass each position, and is information that defines or controls the vehicle's movement, i.e., the time and the position it will be at at that time. The target trajectory can decelerate the vehicle by specifying the time and position. The target trajectory can delay the start timing of a lane change by specifying the time and position. Specifically, by specifying the time and position, the target trajectory can be set to decelerate the vehicle, and after the following vehicle changes lanes in accordance with the predicted trajectory (after the following vehicle has passed), the vehicle can be made to travel on a target trajectory that shares some or all of the same position as the predicted trajectory.

[0017] The autonomous driving function of the processor 10 causes the host vehicle to autonomously drive a target trajectory to a destination based on a driving plan via the vehicle controller 200. The autonomous driving function includes a lane change function that autonomously moves the host vehicle from its own driving lane to an adjacent lane. The driving plan is a group of control instructions that includes a plan for lane changes and command values ​​for autonomously moving the host vehicle to a destination.

[0018] The processor 10 is connected to a communication device 30 including a CAN (Controller Area Network) or other in-vehicle LAN, and exchanges information with each device inside the driving control system 100. The processor 10 receives commands from a user via an input / output device 20 and presents the driving control content as the calculation result to the user. Text, voice, and image data are used for input and output of information.

[0019] The vehicle controller 200 operates based on a group of control commands output from the driving control device 1. The vehicle controller 200 includes a steering control device 210 and a drive control device 220. The vehicle controller 200 acquires command values ​​for autonomous driving control according to the driving plan formulated by the processor 10, and causes the vehicle to travel along a target trajectory on a route to the destination. The target trajectory is calculated based on the predicted trajectory of the following vehicle. The target trajectory is recalculated according to changes in the situation, and a new target trajectory is set. The driving plan is updated according to the new target trajectory, and the command values ​​and control commands for executing the driving plan are also updated.

[0020] The vehicle controller 200 drives the host vehicle according to a set speed. The set speed may be set by the driver using the input / output device 20, or may be automatically set according to predetermined standards based on detected information such as the distance to the preceding vehicle, the relative speed and relative acceleration relative to the preceding vehicle, and other information in accordance with laws and regulations. Command values ​​for driving control are generated by the vehicle controller 200 or the processor 10. The vehicle controller 200 inputs longitudinal and lateral forces that control the traveling position of the host vehicle V1 based on the command values. In accordance with these inputs, the behavior of the vehicle body and the behavior of the wheels are controlled so that the host vehicle autonomously travels along a route to the destination. Based on these controls, at least one of the drive actuator and the brake actuator of the vehicle body drive mechanism controlled by the drive control device 220 and the steering actuator of the steering control device 210, which is activated as needed, operate, thereby executing autonomous driving control that causes the vehicle to autonomously travel along a target trajectory to the destination. The autonomous driving control includes lane keeping control, lane change control, lateral position change control, speed control, and acceleration control. The vehicle controller 200 can also perform driving in accordance with command values ​​based on manual operation by the driver.

[0021] The processing steps of the driving control method executed by the processor 10 will be described with reference to the flowchart of FIG. 2. Based on the detection information from the sensor 2, the processor 10 acquires the current position of the host vehicle using the host vehicle information acquisition device 3 (S1a) and acquires the positions of other vehicles using the other vehicle information acquisition device 4 (S1b). Other vehicles include following vehicles, preceding vehicles, and vehicles traveling in adjacent lanes. The processor 10 references the positions of the other vehicles and the lane information 51 in the map information 5 (S1c), compares the center position of the lane with the position of the other vehicle, and determines that the lane in which the other vehicle is traveling is the lane whose center position is closest to the position of the other vehicle. The processor 10 determines the relationship between the positions of the host vehicle and other vehicles based on the above determination and information (S2). The processor 10 periodically acquires information such as the presence or absence, position (distance), relative speed, and relative acceleration of other vehicles using the detection information from the sensor 2, and sequentially determines the driving conditions, including the positional relationship between the host vehicle and other vehicles, and changes therein (S3). If a following vehicle traveling behind the vehicle in the same lane as the vehicle is recognized (YES in S4), the process proceeds to calculating the predicted trajectory of the following vehicle (S5); otherwise, the search for the following vehicle continues (NO in S4, return to Start).

[0022] The processor 10 determines whether the following vehicle is rapidly approaching the host vehicle (S5). This determination is made by determining whether the following conditions are met: (i) the distance between the host vehicle and the following vehicle traveling behind it is less than a predetermined distance Dmin, and (ii) the acceleration of the following vehicle is higher than the acceleration of the host vehicle. Satisfaction of this condition indicates that the following vehicle is rapidly approaching the host vehicle and the distance between them is decreasing. The predetermined distance Dmin may be set to the distance in the direction of travel of the trajectory reference model under the lane change condition according to the speed, or may be set to a value shorter than a typical inter-vehicle distance (approximately vehicle speed x 1.0 to 1.2 seconds). Additional requirements may include that the relative speed of the following vehicle with respect to the host vehicle is equal to or greater than a threshold Vmax and / or that the acceleration of the host vehicle is equal to or greater than a threshold Amin. Vmax and Amin can be set taking into account measurement error, response delay, etc. The threshold Amin for the host vehicle's acceleration may be a positive number greater than or equal to 0 (i.e., the host vehicle is not decelerating). The distance between the host vehicle and the following vehicle can be obtained from the detection information of the host vehicle's sensor 2. The difference in acceleration between the host vehicle and the following vehicle can be calculated from the speed of the host vehicle and the speed of the following vehicle obtained by the host vehicle's sensor 2. If the condition for determining that the following vehicle is rapidly approaching is not met (NO in S5), a first trajectory is calculated (S6). If this condition is met (YES in S5), it is determined that the following vehicle is rapidly approaching the host vehicle, and a second trajectory is calculated (S8). Until the following vehicle completes a lane change on the first trajectory, the first distance L1 and the second distance L2 on the second trajectory are different, and the second distance L2 is shorter than the first distance L1 (second distance L2 < first distance L1). The first distance L1 and the second distance L2 are the distances in the direction of travel of the predicted trajectory, i.e., along the extension direction of the following vehicle's lane and adjacent lanes.

[0023] Even if it is determined that the above condition is met (Yes in S5), if the processor 10 determines that the host vehicle is decelerating, that is, the acceleration of the host vehicle is negative, based on the host vehicle information obtained from the host vehicle information acquisition device 3 (Yes in S10), the processor 10 proceeds to S6 and calculates the first trajectory as the predicted trajectory of the following vehicle. This process makes it possible to eliminate cases where the acceleration of the following vehicle is higher than the acceleration of the host vehicle and this is due to the deceleration of the host vehicle. This makes it possible to calculate the second trajectory only when the following vehicle is rapidly approaching.

[0024] FIG. 3(a) illustrates an example of the first trajectory TJ21. The first trajectory TJ21 is calculated when the above conditions are not satisfied and is a predicted trajectory of the following vehicle V2 traveling behind the host vehicle V1. The processor 10 calculates movement areas TJ21a, TJ21b, TJ21c, and TJ21d in which the following vehicle V2 exists at multiple times over time, and calculates the first trajectory TJ21 that passes through those areas. The distance along the traveling direction P of the first trajectory TJ21 is the first distance L1. The first distance L1 can be calculated as [L1 = speed Vr of the following vehicle × (road width W / lateral speed of lane change V*lat1)]. The lateral speed of lane change V*lat1 can be set to a typical value (approximately 0.5 to 0.75 m / s).

[0025] FIG. 3(b) illustrates an example of the second trajectory. The second trajectory TJ22 is calculated when the above conditions are met and is a predicted trajectory of the following vehicle V2 traveling behind the host vehicle V1. The processor 10 calculates movement areas TJ22a, TJ22b, TJ22c, and TJ22d in which the following vehicle V2 exists at multiple times over time, and calculates the second trajectory TJ22 that passes through those areas. The distance along the traveling direction of the second trajectory TJ22 is the second distance L2. The second distance L2 can be calculated as [L2 = speed Vr of the following vehicle × (road width w / lateral speed V*lat2 of lane change)]. The lateral speed V*lat2 of the lane change can be set to a speed higher than the lateral speed V*lat1 (approximately 1.0 m / s). The speed of the following vehicle is measured by the sensor 2. The road width may be acquired based on an image captured by the camera of the sensor 2, or may be acquired from the map information 5 or lane information 51. As shown in Figures 3(a) and 3(b), the predicted trajectories TJ21 and TJ22 include a lane change trajectory that involves moving laterally to an adjacent lane. In this embodiment, when a following vehicle V2 traveling behind the host vehicle V1 is recognized, it is assumed that the following vehicle V2 will change lanes, and predicted trajectories TJ21 and TJ22 of the following vehicle V2 are calculated to prepare for the following vehicle V2's lane change. As a result, even if the following vehicle V2 starts changing lanes at the same time as the host vehicle V1, the host vehicle V1 can change lanes without interfering with the following vehicle V2. As a result, it is possible to prevent the following vehicle V2 and the host vehicle V1 from coming close to each other.

[0026] The processor 10 realizes a target trajectory calculation function for the host vehicle. The processor 10 calculates a target trajectory for the host vehicle based on the calculated first trajectory TJ21 of the following vehicle (S7), and calculates a target trajectory for the host vehicle based on the second trajectory TJ22 (S9). In calculating the target trajectory, the processor 10 calculates the target trajectory so that the movement area of ​​the host vehicle V1 traveling on the target trajectory does not interfere with the movement area (TJ21a, TJ21b, TJ21c, TJ21d, etc.) of the following vehicle V2 traveling on the first trajectory or the second trajectory. Note that a method for calculating the target trajectory of the host vehicle that avoids interference with the predicted trajectories and movement areas of other moving bodies can use methods known at the time of filing this application.

[0027] 4(a) to 4(d) show examples of target trajectory TJ1. The direction of travel in each diagram is indicated by an arrow P. Here, the target trajectory TJ1 is simply shown by the initial movement direction. However, as a trajectory, a movement area of ​​the host vehicle corresponding to the movement area of ​​the following vehicle (TJ21a, TJ21b, TJ21c, TJ21d, etc.) shown in FIG. 3 is generated, and the target trajectory TJ1 passing through these areas is calculated. In FIG. 4(a), the following vehicle V2 (T1) at time T1 is predicted to travel along predicted trajectories TJ21 and TJ22, change lanes to an adjacent lane, and move toward the following vehicle V2 (T2) at time T2 (later than T1). In this situation, if the following vehicle V1 also plans to change lanes, and the movement area of ​​the predicted trajectories TJ21 and TJ22 of the following vehicle overlaps with the movement area of ​​the target trajectory, the lane change along the target trajectory is temporarily suppressed. For example, the timing to start changing lanes is delayed by decelerating the host vehicle V1. Even if the target trajectory TJ1 of the host vehicle V1 and the predicted trajectory TJ2 of the following vehicle V2 share some or all of the same positions, the delay in the timing to start changing lanes delays the timing at which the host vehicle V1 passes relative to the timing at which the following vehicle V2 passes each position on the predicted trajectory TJ2, thereby making it possible to calculate a target trajectory TJ1 in which the movement area of ​​the target trajectory TJ1 of the host vehicle V1 does not interfere with the movement area of ​​the predicted trajectory TJ2 of the following vehicle V2.

[0028] FIG. 4(b) shows a situation in which another vehicle V3 (T1) ahead of the host vehicle V1 (T1) decelerates. The following vehicle V2 (T1) changes lanes in the same manner as in FIG. 4(a). The processor 10 calculates a target path TJ1 for moving the host vehicle V1 from the position of the host vehicle V1 (T1) to the position of the host vehicle V1 (T2). Generally, when the preceding another vehicle V3 decelerates, the host vehicle V1 is forced to decelerate as well. The host vehicle V1 (T1) controls the longitudinal position of the host vehicle V1 (T2) at timing T2 by suppressing deceleration so as not to approach the following vehicle V2. At the same time, the lateral position (position in the direction of road width W) of the host vehicle V1 (T2) is shifted to the left in the traveling direction P with respect to the center line K so as not to interfere with the movement area of ​​the following vehicle V2 (see FIGS. 3(a) and 3(b)). The shift direction is a direction that separates the host vehicle V1 from the predicted trajectories TJ21, TJ22 of the following vehicle V2. The host vehicle V1 (T1) moving along the target route TJ1 does not interfere with the movement area of ​​the host vehicle V1 (T2) and the movement area of ​​the following vehicle V2, and the following vehicle V2 can easily overtake the host vehicle V1.

[0029] FIG. 4(c) shows a scene in which another vehicle V3 traveling in the adjacent lane L1 moves along a trajectory TJ3 and approaches the host vehicle V1. At the same timing T1, the following vehicle V2 (T1) is predicted to change lanes along predicted trajectories TJ21 and TJ22, as in FIG. 4(a). If there is no following vehicle, the host vehicle V2 simply moves laterally to the dashed-line frame V1' on the right side of the traveling direction P to move away from the other vehicle V3. However, if the following vehicle V2 is present, the processor 10 calculates a target trajectory that suppresses lateral movement of the host vehicle V1 to the right and guides the host vehicle V1 to a lateral position (V1) that does not overlap with the movement area of ​​the following vehicle. When the following vehicle V2 is present, the target trajectory guides the host vehicle V1 to a position that is farther away from the predicted trajectories TJ21 and TJ22 of the following vehicle V2 and closer to the other vehicle V3 than the position V1' when the following vehicle V2 is not present (a position shifted to the left with respect to the traveling direction P). This allows the host vehicle V1 to maintain a distance from the other vehicle V3 on its side, while also maintaining a distance from the following vehicle V2 that is overtaking the host vehicle V1 along the predicted trajectories TJ21 and TJ22. The host vehicle V1 does not become an obstacle to the other vehicle V3, and does not become an obstacle to the following vehicle V2. This makes it easier for the following vehicle V2 to overtake the host vehicle V1.

[0030] In FIG. 4(d), the following vehicle V2 (T1) changes lanes along predicted trajectories TJ21 and TJ22, as in FIG. 4(a). The processor 10 determines the lateral position of the following vehicle V2 relative to the lane L2 on which the following vehicle V2 is traveling based on other vehicle information, and calculates a target trajectory TJ1 for moving the host vehicle V1 to an adjacent lane on the side away from the following vehicle V2 based on the lateral position. In FIG. 4(d), the position of the following vehicle V2 along the road width direction W in the traveling lane L2 has shifted to the right in the traveling direction P with respect to the lane center line K. In other words, the lateral position of the following vehicle V2 in the lane L2 has shifted toward the adjacent lane L3. The processor 10 calculates a target trajectory TJ1 for moving the host vehicle V1 to the adjacent lane L1 on the opposite side (to the left in the traveling direction P) so as to move away from the following vehicle V2. The host vehicle V1 does not obstruct the lane change of the following vehicle V2.

[0031] Returning to S13 in Fig. 2, the processor 10 activates an autonomous driving function including autonomous lane change control, and controls the host vehicle V1 via the vehicle controller 200 so that it travels along a target trajectory. The processor 10 calculates the acceleration / deceleration and steering angle command value of the host vehicle V1 so that the difference between the actual trajectory of the host vehicle V1 and the target trajectory TJ1 is minimized while preventing the movement area of ​​the target trajectory TJ1 of the host vehicle V1 from overlapping with the movement area of ​​the predicted trajectory TJ2 of the following vehicle V2. This type of control can be realized, for example, by an obstacle avoidance algorithm such as optimal control using a potential function based on the distance to the obstacle.

[0032] In avoidance control, which allows a host vehicle to avoid approaching other vehicles, it is known that other vehicles in front or to the side can be avoided by steering or braking. However, when approaching from behind, avoidance by accelerating is difficult due to constraints such as speed limits. While lateral movement, such as lane changes, can also be considered to avoid approaching a following vehicle, if the host vehicle changes lanes at the same time as the following vehicle, the approaching state between the two vehicles cannot be resolved. In particular, avoiding a following vehicle rapidly approaching the host vehicle is not easy. In this embodiment, when a following vehicle V2 is recognized within a predetermined range, the following vehicle V2 is assumed to be changing lanes and its movement is predicted. The predetermined range can be defined by a distance based on the current position of the host vehicle V1, depending on the speed of the host vehicle V1. As a result, even if the following vehicle V2 begins to change lanes at the same time as the host vehicle V1 changes lanes, a target trajectory TJ1 that maintains a distance from the following vehicle V2 according to the situation can be calculated to avoid the following vehicle V2. Furthermore, when the following vehicle V2 rapidly approaches the host vehicle V1, the target trajectory TJ1 is determined by calculating a predicted trajectory TJ2 with a large lateral position change rate. This prevents the occurrence of an approaching state even when the following vehicle V2 actively approaches the host vehicle V1 and changes lanes.

[0033] The lane change trajectory of the following vehicle V2 may vary depending on the driver's characteristics. Some drivers may close the distance to the host vehicle V1 ahead and then quickly (at a high steering speed) change lanes. If a typical predicted lane change trajectory is applied to such a following vehicle V2, it is likely to overlap with the target trajectory of the host vehicle, resulting in unnecessary avoidance maneuvers. In this embodiment, conditions are set to determine whether the following vehicle V2 is approaching rapidly, and situations in which the following vehicle V2 actively approaches rapidly toward the host vehicle V1 are extracted. Then, different predicted trajectories TJ21 and TJ22 are calculated when the conditions are met and when the conditions are not met. When the following vehicle V2 is not approaching rapidly (when the conditions are not met), a first trajectory TJ21 of a typical lane change is calculated. When the following vehicle V2 is approaching rapidly (when the conditions are met), a second trajectory TJ22 is calculated, which has a shorter distance to complete the lane change than the first trajectory TJ21 (L2<L1). The second trajectory TJ22 is a trajectory along which the rapidly approaching vehicle V2 closes the distance to the host vehicle V1 and prepares for a lane change at a relatively high lateral speed. Because the target trajectory TJ1 is calculated based on the predicted trajectories TJ21 and TJ22, the host vehicle V1 can avoid the following vehicle V2 using an appropriate target trajectory TJ1 according to the situation. This driving control method eliminates the close proximity state between the following vehicle V2 and the host vehicle V1, suppresses the occurrence of the close proximity state, and prevents unnecessary avoidance operations from being performed. As a result, the host vehicle V1 can travel with a sense of security, and prevents passengers of the host vehicle V1 and the following vehicle V2 from feeling uneasy.

[0034] Furthermore, in the driving control method of this embodiment, the direction in which the predicted trajectory will extend is predicted in order to appropriately calculate the predicted trajectory of the following vehicle. The first prediction process will be described below with reference to FIG. 5. In the first prediction process, the direction in which the predicted trajectory will extend is predicted based on the lane conditions. The processor 10 determines whether there is an adjacent lane into which the following vehicle can change lanes based on information detected by the sensor 2 and / or the acquired lane information 51 or the lane information 51 in the map information 5. As shown in FIGS. 5(a) and 5(b), if it is determined that there is an adjacent lane L1 or L2 on either the left or right of the following vehicle V2 traveling in lane L1 or L2, the processor 10 calculates a predicted trajectory TJ2 for causing the following vehicle V1 to change lanes into the adjacent lane L1 or L2. The predicted trajectory TJ2 includes a first trajectory TJ21 and a second trajectory TJ22. In the example of Figure 5(a), the predicted trajectory TJ2 of the following vehicle V2 is predicted to extend into adjacent lane L2, and in the example of Figure 5(b), the predicted trajectory TJ2 of the following vehicle V2 is predicted to extend into adjacent lane L1. In the examples shown in Figures 5(c) and 5(d), lane L4 is the overtaking lane PL. As shown in the figures, if it is determined that there are adjacent lanes (L2 and L4 in Figure 5(c), L1 and L3 in Figure 5(d)) on both the left and right sides of the following vehicle V2's driving lane (L3 in Figure 5(c), L2 in Figure 5(d)), the adjacent lane on the overtaking lane PL (L4) side is selected, and a predicted trajectory TJ2 is calculated to cause the following vehicle V2 to change lanes into the selected adjacent lane. In the situation shown in FIG. 5(c), the predicted trajectory TJ2 of the following vehicle V2 is predicted to extend into adjacent lane L4. In the situation shown in FIG. 5(d), the predicted trajectory TJ2 of the following vehicle V2 is predicted to extend into adjacent lane L3. Processor 10 may determine the passing lane PL in accordance with the laws and regulations of each country or region (in Japan, the rightmost lane of multiple lanes is the passing lane), or it may compare the vehicle speeds of vehicles traveling in each lane detected by sensor 2 (which may be a representative value such as the average speed of multiple vehicles) and determine the lane with the relatively higher vehicle speed as the passing lane. In this way, the adjacent lane into which following vehicle V2 will move can be selected based on the lane conditions and lane attributes. By improving the prediction accuracy of the lane change direction (left or right) of following vehicle V2, the predicted trajectory TJ2 of the following vehicle can be accurately calculated, and the target trajectory TJ1 based on this can be used to cause the host vehicle V1 to take appropriate evasive action. As a result, the sense of security of the user of the vehicle V1 can be improved.Furthermore, appropriate avoidance actions by the host vehicle V1 can also improve the sense of security of the driver of the following vehicle V2.

[0035] The second prediction process will be described with reference to FIG. 6. In the second prediction process, the direction of the predicted trajectory is predicted based on the status of other vehicles around the host vehicle. In this example, it is assumed that there is an adjacent lane into which the following vehicle V2 can change lanes. The determination of the adjacent lane is the same as in the first prediction process. The prediction process shown in FIG. 6(a) focuses on the presence of other vehicles V3 and V4. The processor 10 determines whether another vehicle is present in either the left or right adjacent lane based on information acquired from the other vehicle information acquisition device 4. In the scene shown in FIG. 6(a), the processor 10 determines that there is another vehicle V3 traveling in the lane L1 adjacent to the lane L2 in which the host vehicle V1 and the following vehicle V2 are traveling. If it is determined that there is another vehicle V3, the processor 10 selects the other adjacent lane L3 in which there is no other vehicle, predicts that there is a high possibility that the following vehicle V2 will change lanes into the selected adjacent lane L3, and calculates a predicted trajectory TJ2 for moving the following vehicle V2 into the adjacent lane L3. Based on the presence of other vehicles V3 and V4, the direction in which the predicted trajectory of the following vehicle V2 will proceed can be predicted with high accuracy.

[0036] The prediction process shown in FIG. 6(b) focuses on the positions of the other vehicles V3 and V4. Based on information acquired from the other vehicle information acquisition device 4, the processor 10 determines whether the other vehicles V3 and V4 are present in the adjacent lanes L1 and L2 on both sides of the lane L2 in which the host vehicle V1 and the following vehicle V2 are traveling, as shown in FIG. 6(b). If the processor 10 determines that the other vehicles V3 and V4 are present, the processor 10 selects the adjacent lane L1 in which the other vehicle V3 is traveling, which has a relatively long distance (D2>D1) from the distances D1 and D2 between the following vehicle V2 and the other vehicles V3 and V4. The processor 10 predicts that the following vehicle V2 is likely to change lanes into the selected adjacent lane L1 and calculates a predicted trajectory TJ2 for moving to the adjacent lane L1. Based on the determination based on the positions of the other vehicles V3 and V4, the direction of the predicted trajectory of the following vehicle V2 can be predicted with high accuracy.

[0037] The prediction process shown in FIG. 6(c) focuses on the speeds of the other vehicles V3 and V4. As shown in FIG. 6(c), the processor 10 acquires the speeds of the other vehicles V3 and V4 traveling in adjacent lanes L1 and L3, where the following vehicle V2 can change lanes, from the other vehicle information acquisition device 4. If the speed Vs4 of the other vehicle V4 is greater than the speed Vs1 of the host vehicle V1 by a predetermined value α or more (Vs4≧Vs1+α), the processor 10 predicts that the following vehicle V2 will change lanes to the adjacent lane L3 in which the other vehicle V4 is traveling, and calculates a predicted trajectory TJ2 for moving to the adjacent lane L3. Because the speed Vs3 of the other vehicle V3 is lower than the speed Vs1 of the host vehicle V1 (Vs1>Vs3), the processor 10 predicts that the following vehicle V2 will not change lanes to the adjacent lane L1 in which the other vehicle V3 is traveling. In this way, this method determines that there is a high possibility that the following vehicle V2 will move into the adjacent lanes L1 and L3 when other vehicles V3 and V4 pass by at high speed. Based on the speeds of the other vehicles V3 and V4, the direction of the predicted trajectory of the following vehicle V2 can be predicted with high accuracy.

[0038] When a following vehicle V2 overtakes a preceding vehicle (host vehicle V1), the following vehicle V2 is likely to select a lane that allows it to travel at a higher speed. This driving control method predicts that lanes in which the following vehicle V2 is likely to move include lanes with no other vehicles, lanes with wide space between other vehicles V3, V4, etc., ahead, and lanes in which other vehicles V3, V4 are traveling at a higher speed even if they are present. Based on the presence and behavior of other vehicles V3, V4 in the adjacent lanes, the system can appropriately select the adjacent lane into which the following vehicle V2 is predicted to change lanes. By improving the accuracy of estimating the direction (left or right) in which the following vehicle V2 will change lanes, the path (extension direction) of the predicted trajectory TJ2 can be quickly and accurately calculated. The target trajectory TJ1 calculated based on the accurate predicted trajectory TJ2 allows the host vehicle V1 to take appropriate evasive action, improving the sense of security of the occupants of the host vehicle V1. Appropriate avoidance behavior of the own vehicle V1 also improves the sense of security of the driver of the following vehicle V2.

[0039] Returning to FIG. 2 , the steps of the first and second prediction processes (S11-S12) will be described. Because the execution of the first and second prediction processes is optional (not required), the flow arrows are indicated by dashed lines. As previously described, after the presence of the following vehicle V2 is recognized in S4 of FIG. 2 (YES in S4), it is determined whether the following vehicle V2 is rapidly approaching the host vehicle V1 (S5). Prediction of the direction of the predicted trajectory can be applied regardless of whether a rapid approach is occurring, i.e., in the calculation process of both the first and second predicted trajectories. Whether the following vehicle V2 is rapidly approaching the host vehicle V1 (YES in S5) or not (NO in S5), the process proceeds to S11. This is because predicting the extension direction is meaningful for both the first and second trajectories. In the first prediction process that considers lane attributes, the processor 10 acquires lane information of the adjacent lanes required for the first prediction process and determines the status of the adjacent lanes (S11). Then, the processor 10 extracts adjacent lanes into which the following vehicle V2 can change lanes, and if there are multiple adjacent lanes, selects an adjacent lane into which the following vehicle V2 is predicted to move (e.g., an adjacent lane on the overtaking lane side) (S12). In a second prediction process that takes into account the presence and behavior of other vehicles, the processor 10 acquires driving information of the other vehicles V3 and V4 necessary for the second prediction process and determines the driving conditions, such as the positions and speeds of the other vehicles V3 and V4 (S11). The processor 10 then selects, as an adjacent lane into which the following vehicle V2 can change lanes, an adjacent lane into which the other vehicles V3 and V4 are not present, an adjacent lane into which the other vehicle V3 is present at a position relatively far from the following vehicle V2, or an adjacent lane into which the other vehicle V4 is traveling at a vehicle speed Vs4 that is higher than the vehicle speed Vs1 of the host vehicle V1 by at least a predetermined value α, as an adjacent lane into which the following vehicle V2 is predicted to move (S12). Once an adjacent lane is selected by the first or second prediction process, the processor 10 obtains identification information for the selected adjacent lane and, as described above, calculates the first trajectory TJ21 and the second trajectory TJ22, which include the trajectory for moving to this adjacent lane (S6, S8).

[0040] Next, a third prediction process will be described with reference to FIGS. 7(a) and 7(b). The third prediction process is a method for selecting the current lane rather than selecting either the left or right adjacent lane as in the first or second prediction process. The processor 10 determines a situation in which the following vehicle V2 cannot change lanes and predicts that the following vehicle V2 will maintain the current lane (will not change lanes). As shown in FIG. 7(a), the following vehicle V2 is traveling behind the host vehicle V1 traveling in lane L2, and an adjacent lane L1 is located to the left of the host vehicle V1 in the direction of travel P of lane L2. However, a boundary line X indicating a lane entry prohibition (lane change prohibition) is located between the adjacent lane L1 and the host vehicle V2. In other words, even if the adjacent lane L1 physically exists, the following vehicle V2 cannot move into the adjacent lane L1 due to the regulations. If there is no lane into which the following vehicle V2 can change lanes, the processor 10 calculates a predicted trajectory TJ2 for the following vehicle V2 to maintain the lane L2 it is currently traveling in. This makes it possible to extract situations in which the following vehicle V2 cannot change lanes due to lane conditions and regulations, and to quickly calculate a predicted trajectory for maintaining the lane.

[0041] Figure 7(b) is similar to Figure 7(a), where a boundary line X that prohibits lane changes is provided for the following vehicle V2, and the following vehicle V2 cannot move into lane L1. Also, based on the detection information of sensor 2 and the determination of the other vehicle information acquisition device 4, if it is determined that the following vehicle V2 does not meet the predetermined lane change conditions, the following vehicle V2 cannot move into the adjacent lane L2 either. The predetermined lane change conditions are predefined from conditions such as whether there is space for the following vehicle V2 to enter the adjacent lane L2, whether the distance from the other vehicle V3 traveling in the adjacent lane L2 is greater than or equal to a predetermined value, whether the relative speed is greater than or equal to a predetermined value (positive), and whether the relative acceleration is greater than or equal to a predetermined value (positive). As the lane change conditions for the following vehicle V2, the lane change conditions of the host vehicle V1 may be applied. The threshold values in the lane change conditions can be set to different values for the host vehicle V1 and the following vehicle V2. Fig. 8 shows an example of lane change conditions based on the relationship between relative speed and distance. When the other vehicle traveling in the adjacent lane belongs to the range of y≧f1(x) or y≦f2(x), it is determined that the lane change conditions are satisfied (lane change is possible), and when the other vehicle traveling in the adjacent lane belongs to the range of f2(x)<y<f1(x), it is determined that the lane change conditions are not satisfied (lane change is not possible). For example, as shown in Fig. 7(b), when the other vehicle V3 is traveling on the side of the following vehicle V2 (the vertical position difference is less than a predetermined value), it is determined that the predetermined lane change conditions are not satisfied. When the following vehicle V2 does not meet the lane change conditions, the processor 10 calculates a predicted trajectory TJ2 for the following vehicle V2 to maintain the lane L2 on which it is currently traveling. As shown in Figs. 7(a) and (b), the change in the lateral position of the calculated predicted trajectory TJ2 is less than or equal to a predetermined amount or zero. By calculating the predicted trajectory TJ2 of the following vehicle V2 and the target trajectory TJ1 of the host vehicle V1 on the premise that the recognized following vehicle V2 will change lanes, the approaching state between the host vehicle V1 and the following vehicle V2 during lane change is eliminated. However, in order to execute a lane change, various prerequisite conditions such as ensuring space after the lane change and ensuring a sufficient inter-vehicle distance are weighted. In this case, if the distance from the following vehicle V2 is not sufficiently separated, the host vehicle V1 cannot change lanes, and there is a possibility that excessive suppression will occur in the driving control of the host vehicle V1. When the following vehicle V2 cannot change lanes, by determining that the following vehicle V2 will maintain its lane, the host vehicle V1 can find an opportunity to execute a lane change.

[0042] Returning to FIG. 2 , the processing steps (S21-S23) of the third prediction process will be described. The processor 10 determines whether the following vehicle V2 can change lanes. If a lane change is not possible, the processor 10 calculates a predicted trajectory that will maintain the current lane. This process can be performed at any time before the predicted trajectory calculation process (S6 and S7). Since this process is optional (not required), the flow arrows are indicated by dashed lines. In FIG. 2 , this process is performed after the following vehicle V2 is recognized (YES in S4). Once the following vehicle V2 is recognized (YES in S4), information about the adjacent lane is acquired (S21). It is determined whether the adjacent lane is a lane where lane changes are prohibited. The processor 10 then reads the lane change conditions described above (S22) and determines whether the following vehicle's lane change conditions are satisfied (S23). If it is determined that the lane change conditions are met (YES in S23), it is determined that the following vehicle V2 is approaching rapidly (S5), and a first trajectory or a second trajectory is calculated (S6, S8). On the other hand, if it is determined that the lane change conditions of the following vehicle V2 are not met (NO in S23), it is determined that the following vehicle V2 will not change lanes. Then, a predicted trajectory TJ2 for maintaining the driving lane is calculated (S24). The processor 10 calculates the trajectory of the following vehicle V2 moving in the current lane as a predicted trajectory TJ2 (first trajectory TJ21 or second trajectory TJ22) (S6, S8), and calculates a target trajectory TJ1 for the calculated predicted trajectory TJ2 (S7, S9).

[0043] 100... driving control system, 1... driving control device, 10... processor, 11... CPU, 12... ROM, 13... RAM, 20... input / output device, 30... communication device, 2... sensor, 3... vehicle information acquisition device, 4... other vehicle information acquisition device, 5... map information, 51... lane information, 6... navigation device, 200... vehicle controller, 210... steering control device, 220... drive control device

Claims

1. A driving control method used in a processor to control autonomous driving of a host vehicle to travel along a target trajectory, wherein the processor: recognizes a following vehicle traveling behind the host vehicle based on host vehicle information and other vehicle information acquired using a sensor; determines whether a condition is satisfied or not that a distance between the following vehicle and the host vehicle is less than a predetermined distance and that the acceleration of the following vehicle is higher than the acceleration of the host vehicle; if it is determined that the condition is not satisfied, calculates a first trajectory in which a distance along the traveling direction of a predicted trajectory until the following vehicle completes a lane change is a first distance; if it is determined that the condition is satisfied, calculates a second trajectory in which a distance along the traveling direction of a predicted trajectory until the following vehicle completes a lane change is a second distance that is shorter than the first distance; and calculates the target trajectory so that a movement area of ​​the host vehicle traveling along the target trajectory does not interfere with a movement area of ​​the following vehicle traveling along the first trajectory or the second trajectory.

2. A driving control method as described in claim 1, wherein the processor calculates the first trajectory as a predicted trajectory of the following vehicle when it is determined that the vehicle is decelerating based on the vehicle information.

3. A driving control method as described in claim 1 or 2, wherein the processor acquires lane information detected by a sensor or lane information from map information, determines whether there is an adjacent lane into which the following vehicle can change lanes, and if it is determined that there is an adjacent lane on either the left or right side, calculates the predicted trajectory for the following vehicle to change lanes to the adjacent lane on that side, and if it is determined that there are adjacent lanes on both the left and right sides, selects the adjacent lane on the overtaking lane side, and calculates the predicted trajectory for the following vehicle to change lanes to the selected adjacent lane.

4. A driving control method as described in claim 1 or 2, wherein the processor acquires lane information detected by the sensor or lane information from map information, and determines whether there is an adjacent lane in which the following vehicle can change lanes; if it is determined that there is another vehicle traveling in the adjacent lane on either the left or right, selects the other adjacent lane in which the other vehicle is not present, and calculates the predicted trajectory for the following vehicle to change lanes to the selected adjacent lane; if it is determined that there is another vehicle traveling in the adjacent lane on both the left and right sides, selects the adjacent lane in which the other vehicle that is relatively far away from the following vehicle is traveling, and calculates the predicted trajectory for the following vehicle to change lanes to the selected adjacent lane.

5. A driving control method as described in claim 4, wherein the processor calculates the predicted trajectory for the following vehicle to change lanes into the adjacent lane in which the other vehicle is traveling when the speed of the other vehicle traveling in the adjacent lane in which the following vehicle can change lanes is greater than a predetermined value higher than the speed of the subject vehicle.

6. A driving control method according to any one of claims 1 to 5, wherein the processor calculates the predicted trajectory such that the following vehicle will maintain the lane in which it is currently traveling when there is no lane in which the following vehicle can change lanes.

7. A driving control method as described in any one of claims 1 to 5, wherein the processor calculates the predicted trajectory for the following vehicle to maintain the lane in which it is currently traveling when it is determined that the following vehicle does not satisfy a specified lane change condition.

8. A driving control method as described in any one of claims 1 to 7, wherein the processor determines the lateral position of the following vehicle relative to the lane in which the following vehicle is traveling based on the other vehicle information, and calculates the target trajectory for moving the vehicle to an adjacent lane to the lane farther from the following vehicle based on the lateral position.

9. A driving control device comprising: a processor that controls autonomous driving to make a host vehicle travel along a target trajectory, wherein the processor recognizes a following vehicle traveling behind the host vehicle based on host vehicle information and other vehicle information acquired using a sensor, and determines whether a condition is satisfied or not that a distance between the following vehicle and the host vehicle is less than a predetermined distance and that the acceleration of the following vehicle is higher than the acceleration of the host vehicle; if it is determined that the condition is not satisfied, calculates a first trajectory in which a distance along the traveling direction of a predicted trajectory until the following vehicle completes a lane change is a first distance; if it is determined that the condition is satisfied, calculates a second trajectory in which a distance along the traveling direction of a predicted trajectory until the following vehicle completes a lane change is a second distance that is shorter than the first distance; and calculates the target trajectory so that a movement area of ​​the host vehicle traveling along the target trajectory does not interfere with a movement area of ​​the following vehicle traveling along the first trajectory or the second trajectory.

Images