Operation control method and operation control device

Abstract

To make an own vehicle follow a target track so that a lateral G jerk does not become large even when the own vehicle deviates from the target track.SOLUTION: The processor 10 acquires the own vehicle information including the current position of the own vehicle V1, calculates the first target trajectory R1 on which the own vehicle travels, calculates the central trajectory MT passing through the center of the traveling road boundaries LL and LR determined based on the traveling road information of the traveling road RT on which the own vehicle V1 travels when it is determined that the own vehicle R1 has deviated from the first target trajectory based on the deviation amount Ld of the lateral position of the own vehicle V1 with respect to the first target trajectory V1 based on the own vehicle information, and calculates the shift amount Lc of the lateral position of the own vehicle wall with respect to the central trajectory MT based on the own vehicle information. V1 R1, when the deviation amount Ld is larger than the shift amount Lc, a differential between the deviation amount Ld and the shift amount Lc is calculated every time, either the central trajectory MT or the corrected trajectory AT passing through a position between the central trajectory MT and the first target trajectory R1 is calculated as the second target trajectory R2 according to the differential, and the own vehicle R2 is driven along the second target trajectory V1.SELECTED DRAWING: Figure 2

Description

[Technical Field]

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

[0002] A technology is known in which an offset amount of the coordinate value or curvature is calculated for each basic orbit coordinate data of each coordinate, which is ordered so that the arc length from the origin increases, and a vehicle is controlled based on target orbit coordinate data to which the offset amount has been added for each basic orbit coordinate data (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. WO2022 / 190910 Summary of the Invention [Problem to be solved by the invention]

[0004] However, if the vehicle deviates from the target trajectory due to an external disturbance, a large steering amount is required to return the vehicle to the target trajectory, which causes a problem of large lateral G jerk in the vehicle.

[0005] The problem to be solved by the present invention is to prevent a large lateral G jerk from occurring in the host vehicle when the host vehicle returns to the target trajectory even if the host vehicle deviates from the target trajectory. [Means for solving the problem]

[0006] The present invention solves the above problem by calculating a second target trajectory, which is a corrected trajectory that passes through the center trajectory or a position between the center trajectory and the first target trajectory, depending on the difference between the deviation amount and the shift amount calculated at a predetermined period, when the amount of deviation of the lateral position of the vehicle from the first target trajectory is greater than the amount of shift of the lateral position of the vehicle from the center trajectory, and driving the vehicle along the second target trajectory. [Effects of the Invention]

[0007] According to the present invention, even if the host vehicle deviates from the target trajectory, it is possible to prevent a large lateral G jerk from occurring in the host vehicle when the host vehicle returns to the target trajectory. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing the configuration of the operation control system. [Figure 2] FIG. 2 is a flowchart showing an example of a procedure for operational control. [Figure 3] FIG. 3 is a diagram showing an example of calculation of the second target trajectory. [Figure 4] FIG. 4 is a functional block diagram of the operation control device. [Figure 5] FIG. 5 is a flowchart showing an example of a procedure for calculating the second target trajectory. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1 shows the hardware configuration of a driving control system 100 equipped with a vehicle driving control device 1 according to this embodiment. A processor 10 of the driving control device 1 uses each piece of hardware in the driving control system 100 to implement this driving control method.

[0010] The driving control system 100 includes one or more sensors 2, a vehicle information acquisition device 3, a road recognition device 4, map information 5, and a navigation device 6. Sensor 2 detects the vehicle and its surroundings. Sensor 2 detects the vehicle's current position (including its relative position to objects), the boundaries of the lane on which the vehicle is traveling, the lane on which the vehicle is traveling, the presence or absence of objects including other vehicles, the distance to the objects, and the relative speed and relative acceleration of the objects. A plurality of sensors 2 are provided on the vehicle, and they work together to form a sensor system. Sensor 2 detects objects including other vehicles traveling in front, behind, on the left and right sides of the vehicle. Sensor 2 provides object detection information to processor 10. Processor 10 determines the object's position, posture, motion state (speed, acceleration, behavior, direction of travel, etc.) and changes therein based on the detection information, and performs autonomous driving control using the determination results. The sensor 2 includes one or more cameras 21 arranged on the vehicle. The cameras 21 capture images of the surroundings of the vehicle in all directions. The cameras 21 include image sensors equipped with imaging elements such as CCDs, ultrasonic cameras, and infrared cameras. The cameras 21 include at least a front camera that captures images in front of the vehicle, a rear camera that captures images behind or on the rear sides of the vehicle, and left and right side cameras that capture images of the left and right sides of the vehicle, and the front and rear of the left and right sides. The type of camera 21 is not limited as long as it can capture images in all directions of the vehicle. The camera 21 of the sensor 2 acquires image information of the road on which the host vehicle V1 is traveling, and based on the image information, detects the left and right boundaries of the road, the road formed between the boundaries, the center of the road, and the position of the host vehicle relative to the road. The sensor 2 includes a radar device 22 that detects (measures distance to) the presence of objects around the host vehicle, as well as their positions and changes in positions. The radar device 22 is a device that measures the distance and direction from the host vehicle to the object, as well as the positional relationship and distance between objects, by emitting electromagnetic waves toward the object and measuring the reflected waves. The radar device 22 includes laser radar, millimeter-wave radar (LRF), a LiDAR unit (light detection and ranging unit), ultrasonic radar, and sonar. The radar device 22 of the sensor 2 acquires measurement information regarding the structure of the road on which the host vehicle V1 is traveling, and based on the measurement information, detects the left and right boundaries of the road, the road formed between the boundaries, the center of the road, and the position of the host vehicle relative to the road. The sensor 2 includes a receiver for signals from a GPS (Global Positioning System) or a GNSS (Global Navigation Satellite System), a gyro sensor or an IMU (Inertial Measurement Unit), and a vehicle speed sensor, and uses these to detect the current position of the vehicle. The sensor 2 can also acquire detection information from each in-vehicle device or external device. Each sensor 2 transmits the acquired detection information to the host vehicle information acquisition device 3, the lane recognition device 4, or the processor 10 in response to a request or command. The processor 10 may acquire the detection information directly from the camera 21 or the radar device 22, or may acquire the detection information via the host vehicle information acquisition device 3 or the lane recognition device 4.

[0011] The host vehicle information acquisition device 3 calculates the current position, attitude, speed, acceleration, behavior, and traveling direction of the host vehicle based on the detection information acquired from the sensor 2, and provides the calculated information to the processor 10. The host vehicle information acquisition device 3 calculates the lateral position of the host vehicle relative to the road, and provides the calculated information to the processor 10.

[0012] The lane recognition device 4 recognizes lane boundaries of the lane on which the vehicle is traveling. The lane boundaries are boundaries in the road width direction of the road surface of the lane on which the vehicle is traveling, and include a right boundary indicating the right edge of the road surface and a left boundary indicating the left edge of the road surface. The lane recognition device 4 recognizes positions where lane marks that separate the lane are present as lane boundaries based on detection information including images captured by the camera 21. The lane recognition device 4 recognizes positions where the road surface is discontinuous as lane boundaries based on detection information including measurement information from the radar device 22. The lane recognition device 4 recognizes positions where objects such as steps, shoulders, or guardrails exist at the ends of the lane as lane boundaries, for example. The lane recognition device 4 acquires lane information including lane boundary position information. The lane recognition device 4 identifies a pair of lane boundaries based on the positions of both end lane boundaries, and calculates a median trajectory that passes through the center position of the pair of lane boundaries. Information on the median trajectory is provided to the processor 10. The median trajectory is used to calculate the first target trajectory and / or the second target trajectory.

[0013] The driving control system 100 includes map information 5. The map information 5 is high-precision map information used for autonomous driving control and includes lane information 51 that identifies each lane. The lane information 51 includes identification information that identifies each of multiple lanes belonging to a road. The map information 5 is stored in one or more of the ROM 12, the storage device of the navigation device 6, and the storage device of an external server that the processor 10 can access via the communication device 30.

[0014] The driving control system 100 includes a navigation device 6. The navigation device 6 refers to map information 5 and calculates a route from the current position to a set destination. The route and target trajectory calculated by the navigation device 6 are provided to the vehicle controller 200 and used for autonomous driving control.

[0015] The driving control system 100 includes a vehicle controller 200. The vehicle controller 200 is equipped with a steering control device 210 and a drive control device 220, and acquires command values ​​for autonomous driving control according to a driving plan formulated by the processor 10 of the driving control device 1, and controls the host vehicle to travel along a target trajectory. The target trajectory is composed of a series of multiple unit target trajectories. The target trajectory includes command values ​​for driving control associated with each point. The command values ​​for driving control are generated by the vehicle controller 200 or the processor 10. The command values ​​are vehicle control command values ​​for driving the host vehicle along the target trajectory. The command values ​​include a set speed (including an upper limit speed and an applicable speed) for driving the vehicle, and the vehicle controller 200 drives the host vehicle according to the set speed. Based on the command values, the vehicle controller 200 inputs longitudinal and lateral forces to the steering control device 210 and the drive control device 220 to control the driving position of the host vehicle. According to these input command values, the behavior of the body and the behavior of the wheels of the host vehicle are controlled so that the host vehicle travels autonomously along a route to the destination. At least one of the drive actuator and the brake actuator of the drive mechanism of the vehicle body controlled by the drive control device 220 and the steering actuator of the steering control device 210, which is activated as needed, operate autonomously, and autonomous driving control is executed to make the vehicle travel autonomously along a target trajectory. In addition, the vehicle controller 200 can execute driving in accordance with command values ​​based on manual operation by the driver input via the input / output device 20.

[0016] The driving control device 1 included in the driving control system 100 controls autonomous driving, causing the vehicle to travel along a target trajectory. The driving control device 1 of this embodiment calculates the latest target trajectory that the vehicle should currently follow based on detection information acquired every hour at a predetermined cycle in response to changes in the situation. The processor 10 included in the driving control device 1 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 this ROM 12, and a RAM (Random Access Memory) 13 that functions as an accessible storage device. The processor 10 implements this driving control method using each piece of hardware in the driving control system 100.

[0017] The processing procedure of the autonomous driving control of this embodiment will be described based on the flowchart in Fig. 2. The processor 10 acquires host vehicle information including the current position of the host vehicle using the sensor 2 (S1). The processor 10 searches for a route from the current position to the destination by referring to the map information 5 (S2). The route search process may be performed by the navigation device 6. A first target trajectory is calculated based on the route from the current position to the destination (S3). The first target trajectory is a driving trajectory calculated so as to optimize the behavioral quantities of the host vehicle while driving. An optimized trajectory is a trajectory for which the evaluation value of the host vehicle's behavior when traveling on that trajectory is less than a predetermined quantity (including a minimum value or minimum value range; the same applies hereinafter). For example, the optimized trajectory may be calculated as a trajectory in which the lateral G jerk (lateral jerk, jerk) or lateral acceleration is less than a predetermined amount. Alternatively, the optimized trajectory may be calculated as a trajectory in which one or more of the yaw rate, yaw rate deviation, steering angle, steering angle change, steering angle change rate, lateral velocity, lateral acceleration, and lateral G jerk of the host vehicle V1 are less than a predetermined amount. The optimized trajectory may be a trajectory in which the curvature of a curve included in the trajectory is less than a predetermined curvature. The first target trajectory may be calculated based on the vehicle performance of the host vehicle. The trajectory optimization process (algorithm) for autonomous driving can use any method known at the time of filing of this application. One example of the optimization method is the sequential conjugate gradient restoration algorithm (SCGRA). SCGRA is an optimization method that generates a target trajectory for the host vehicle by using the steepest descent method until a constraint condition is satisfied and then performing a convergence calculation based on the conjugate gradient method until the evaluation value of an evaluation function is minimized. The constraint conditions are essential conditions that the target trajectory must satisfy, and the evaluation function is used to evaluate conditions that are important in driving. Furthermore, if the first target trajectory has a straight section and a curved section, a transition curve connecting them can be provided between them. The transition curve can be calculated using a method known at the time of filing, such as a cresoid curve, a sine half-wavelength curve, a cubic parabola, or a hyperdiminished curve. If the first target trajectory includes a curved section, an out-in-out trajectory can be used when entering, traveling through, and exiting the curved section. The target trajectory (including the first target trajectory and the second target trajectory; the same applies below) is defined by multiple pieces of position information. The target trajectory is constructed by connecting multiple unit target trajectories of a predetermined section separated by multiple nodes. Information on each point of the target trajectory is associated with control command values ​​required for driving the vehicle, such as the speed, acceleration, yaw angle, and yaw rate at each point. The first target trajectory can be defined by the amount of lateral shift relative to the center trajectory of the path that constitutes the route to the destination. The processor 10 can specify, as the first target trajectory, a trajectory that passes through each point offset to the right or left from the point that defines the calculated center trajectory. The processor 10 can calculate the amount of lateral shift relative to the center trajectory of the first target trajectory calculated by the optimization calculation, and use the calculated shift amount to define the target trajectory (each point) relative to the center trajectory.

[0018] Processor 10 starts executing driving control to make the host vehicle travel along the first target trajectory (S4). Processor 10 monitors whether the host vehicle is traveling along the first target trajectory (S5). Processor 10 compares the current position of the host vehicle with the position of the first target trajectory, and confirms that the host vehicle is traveling along the first target trajectory. Specifically, processor 10 calculates the amount of deviation of the lateral position of the host vehicle from the first target trajectory based on the host vehicle information (S6). Processor 10 determines whether the host vehicle has deviated from the first target trajectory based on the amount of deviation (S7). Specifically, processor 10 compares the lateral position (position in the vehicle width direction) of the current position of the host vehicle information with the lateral position (position in the road width direction) of the first target trajectory. If the difference is less than a predetermined first deviation threshold, processor 10 determines that the host vehicle is traveling following the first target trajectory. If the difference is equal to or greater than a predetermined second deviation threshold (second deviation threshold ≥ first deviation threshold), processor 10 determines that the host vehicle is not traveling following the first target trajectory and is traveling deviating from the first target trajectory. The first deviation threshold and the second deviation threshold may be the same value or different values. This process continues at a predetermined interval until a deviation is determined (NO in S7, S5-S7). If the traveling environment is ideal for executing driving control, the possibility of the host vehicle deviating from the first target trajectory is extremely low. However, the traveling environment is not always constant, and the traveling trajectory of the host vehicle may deviate from the first target trajectory due to sudden crosswinds, the appearance of obstacles, changes in road surface conditions, etc. The processor 10 determines that the host vehicle has deviated from the first target trajectory without being able to follow the first target trajectory based on the deviation amount of the lateral position of the host vehicle from the first target trajectory. The processor 10 determines the state of deviation of the host vehicle from the first target trajectory (S8). The processor 10 calculates an appropriate target trajectory depending on the state of deviation.

[0019] Processor 10 uses sensor 2 to acquire road information for the road on which the host vehicle is traveling (S9). Processor 10 determines road boundaries based on the road information (S10). The road information includes detection information for the right and left road boundaries of the road. Processor 10 detects the road boundaries based on the road information, including images captured by camera 21 and / or measurement information from radar device 22. A road area on which the host vehicle is traveling is formed between the left and right road boundaries. Processor 10 detects the road boundaries based on the detection information and calculates a center track that passes through the center point of the road boundary along the road width direction (S11). Processor 10 compares the position of the center track with the current position of the host vehicle in the vehicle information and calculates the amount of lateral shift of the host vehicle relative to the center track (S12). The shift amount is the distance between the center track along the road width direction (lateral direction) and the current position of the host vehicle. Processor 10 compares the deviation amount calculated in S6 with the shift amount calculated in S12 (S13) and determines whether the deviation amount is greater than the shift amount (S14). If the deviation amount is less than or equal to the shift amount (NO in S14), the host vehicle is driven along the first target trajectory (S15). In other words, the driving control started in S4 is continued.

[0020] On the other hand, if the deviation amount is greater than the shift amount (YES in S14), a calculation process for a second target trajectory according to the difference is executed (S16). The current position of the traveling vehicle changes from moment to moment, and the deviation amount and shift amount change accordingly, and the difference between them also changes. The processor 10 repeatedly calculates the difference between the deviation amount and the shift amount at a predetermined cycle (S17). The processor 10 calculates a center trajectory or a corrected trajectory based on the newly calculated difference (S18). The corrected trajectory is a trajectory that passes through any position between the center trajectory and the first target trajectory. The corrected trajectory does not include the center trajectory. The processor 10 calculates, as the second target trajectory, either the center trajectory or a corrected trajectory that passes through a position between the center trajectory and the first target trajectory according to the difference between the deviation amount and the shift amount (S19). The second target trajectory may be a corrected trajectory that passes through a position closest to the current position of the vehicle among multiple corrected trajectories. The corrected trajectory is repeatedly calculated based on the newly calculated difference in S17. When the host vehicle deviates from the first target trajectory, the processor 10 sets a second target trajectory calculated according to the difference between the deviation amount and the shift amount based on the host vehicle's current position at the time of the deviation as the target trajectory. The second target trajectory is calculated at a predetermined period based on the difference between the deviation amount and the shift amount based on the host vehicle's current position, which continues to move from the deviation until returning to the first target trajectory, and the calculated second target trajectory is set as the target trajectory. The processor 10 drives the host vehicle along the set second target trajectory (S20).

[0021] In this embodiment, when the host vehicle deviates from the initially set first target trajectory, a second target trajectory is calculated based on the difference between the deviation amount of the lateral position based on the current position of the host vehicle and the shift amount of the lateral position, and the host vehicle is driven to move from the deviation position along the second target trajectory. Specifically, when the deviation amount is greater than the shift amount, the difference between the deviation amount and the shift amount is calculated at a predetermined interval, and a corrected trajectory passing through the center trajectory or a position between the center trajectory and the first target trajectory is calculated based on the difference. As the host vehicle deviates from the first target trajectory and moves along the calculated second target trajectory, the position of the host vehicle is gradually corrected. Furthermore, the second target trajectory calculated according to the host vehicle's position gradually approaches the first target trajectory. By executing driving control to make the host vehicle follow the second target trajectory calculated for each process, the host vehicle can be gradually returned to the first target trajectory. In this way, by executing driving control to make the vehicle follow the second target trajectory rather than performing driving control to return the vehicle to the first target trajectory immediately after deviation, it is possible to prevent a large lateral G jerk from occurring when returning to the first target trajectory, even if the vehicle deviates from the first target trajectory. Incidentally, the inventors applied a disturbance to an experimental vehicle traveling on a predetermined trajectory, which is a target trajectory, to cause the experimental vehicle to deviate to the outside of a turn, and measured the lateral jerk (m / s 3 A simulation was carried out in which the lateral jerk (m / s) was measured over time. In the comparative example, the vehicle was driven with the predetermined trajectory as the target trajectory when it deviated, and in the example, the vehicle was driven with the second target trajectory calculated based on the method of this embodiment as the target trajectory when it deviated. The lateral jerk (m / s) was measured over time from the time the test vehicle deviated until it returned to the predetermined trajectory. 3 The lateral jerk (m / s 3 The maximum and minimum difference (PP) of the lateral jerk (m / s 3 ) was 71.8% of the maximum-minimum difference (PP). It was confirmed that the restoration method of this embodiment suppresses the amount of lateral jerk more effectively than the comparative example.

[0022] In this embodiment, a host vehicle that has deviated from a first target trajectory is controlled to follow a second target trajectory from the position at which the deviation was determined, aiming to return to the first target trajectory. The second target trajectory is a target trajectory set to eliminate the deviation from the first target trajectory. The processor 10 drives the host vehicle to follow the second target trajectory until the host vehicle returns to the first target trajectory (NO in S21) (S20). The second target trajectory is calculated every hour based on the difference between the deviation amount and the shift amount calculated at a predetermined interval, and is set as a new target trajectory for the driving control. The repeatedly calculated second target trajectory R2 gradually approaches the first target trajectory R1. When the processor 10 determines that the second target trajectory overlaps with the first target trajectory (they have a common trajectory) and that the host vehicle has returned to the first target trajectory (YES in S21), the processor 10 drives the host vehicle along the first target trajectory (S22). The processor 10 determines that the vehicle has returned from the second target trajectory to the first target trajectory when the second target trajectory and the first target trajectory overlap, or when the difference in the lateral positions (positions along the width direction of the road) of one or more points between each point dividing the second target trajectory and each point dividing the first target trajectory becomes less than a predetermined value. In this way, even if the host vehicle deviates from the first target trajectory due to a disturbance, the processor 10 periodically calculates the difference between the deviation amount of the lateral position that changes as the host vehicle moves and the shift amount of the lateral position, and repeatedly calculates the second target trajectory based on the calculated difference. The repeatedly calculated second target trajectory gradually approaches the first target trajectory. When the processor 10 determines that the calculated second target trajectory overlaps the first target trajectory, it transitions from driving control that causes the host vehicle to follow the second target trajectory to driving control that causes the host vehicle to follow the first target trajectory. This makes it possible to suppress the steering amount and the amount of lateral G jerk generated during return driving to return the host vehicle from the deviation state to the original first target trajectory.

[0023] Based on FIG. 3, the calculation method of the second target trajectory R2 (shown by the dashed line) will be described. FIG. 3 shows three patterns 1 to 3 of the second target trajectory R2. FIG. 3 shows the host vehicle V1, the lane RT, the right lane boundary LR, the left lane boundary LL, the center trajectory MT, the first target trajectory R1, the second target trajectory R2, the return trajectory RB, the deviation amount Ld, the shift amount Lc, and the lane width Lb. The lane RT is the lane on which the host vehicle V1 travels. The right lane boundary LR and the left lane boundary LL are the boundaries in the lane width direction of the lane. The center trajectory MT is a trajectory passing through the central point in the lane width direction of the lane. The first target trajectory R1 is a trajectory along the route from the current position of the host vehicle V1 to the destination, and is a trajectory optimized for the behavior of the host vehicle V1 during travel. The second target trajectory R2 is a trajectory along which the host vehicle V1 travels to return from the current position of the deviated host vehicle V1 to the first target trajectory V1 when the host vehicle V1 deviates from the first target trajectory V1. The return trajectory RB is a trajectory predicted to be traveled by the deviated host vehicle V1 from the current position to the set second target trajectory. The deviation amount Ld is the distance between the host vehicle V1 determined to have deviated from the first target trajectory V1 and the first target trajectory R1. The shift amount Lc is the distance between the current position of the host vehicle V1 determined to have deviated from the first target trajectory R1 and the center trajectory MT. The distance is the distance along the lane width direction of the lane. In FIG. 3, the lane width direction or the vehicle width direction is shown by the X-axis, and the extending direction of the lane or the vehicle length direction is shown by the Y-axis. The deviation amount Ld and the shift amount Lc are the distances in the X direction with respect to the first target trajectory R1. The deviation amount Ld and the shift amount Lc are absolute values. The lane width Lb is the width of the lane RT. The processor 10 calculates the second target trajectory R2 based on the difference Ld - Lc between the deviation amount Ld and the shift amount Lc.

[0024] As shown in FIG. 3, Pattern 1 is the case where Ld - Lc is zero or less (0 ≧ Ld - Lc), and Patterns 2 and 3 are the cases where Ld - Lc is greater than zero (0 < Ld - Lc). The calculation methods of the second target trajectory in each pattern will be described respectively. <Pattern 1> Pattern 1 shows an example of the second target trajectory R2 calculated when the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is zero or less (0≧Ld-Lc). In pattern 1, the shift amount Lc of the lateral position of the host vehicle V1 relative to the center trajectory MT is greater than the deviation amount Ld of the host vehicle V1 relative to the first target trajectory R1. In other words, the host vehicle V1 that has deviated from the first target trajectory R1 is located closer to the first target trajectory R1 than the center trajectory MT. When the difference between the deviation amount and the shift amount is zero or less (0≧Ld-Lc), the processor 10 calculates the first target trajectory R1 as the second target trajectory R2. The second target trajectory R2 is common to the first target trajectory R1. In addition, K in FIG. 3 is a coefficient indicating the degree of reflection of the first offset amount of the first target trajectory based on the central trajectory MT when calculating the second target trajectory R2. When the host vehicle V1 travels along the first target trajectory R1, it follows the first offset amount of the first target trajectory R1, so the coefficient K is 1 (K=1). In pattern 1, the host vehicle V1 that has deviated laterally from the first target trajectory R1 by the deviation amount Ld is driven autonomously using the first target trajectory R1 as the second target trajectory R2. After traveling along the return trajectory RB, the host vehicle V1 is controlled to travel along the second target trajectory R2 (=first target trajectory R1). The second target trajectory R2 is calculated at a predetermined interval. The second target trajectory R2 calculated at each predetermined interval gradually approaches the first target trajectory. When the processor 10 determines that the second target trajectory R2 overlaps with the first target trajectory R1 and the host vehicle V1 has returned to the first target trajectory R1, the processor 10 drives the host vehicle V1 along the first target trajectory R1. When it is determined that the host vehicle V1 is located close to the first target trajectory R1 during deviation, the driving of the host vehicle V1 is controlled so that it follows the first target trajectory R1 as the second target trajectory R2, so that a large steering amount is not required to eliminate the deviation state, and the amount of lateral jerk caused by the steering amount can be suppressed.

[0025] <Pattern 2> Pattern 2 shows an example of the second target trajectory R2 calculated when the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is equal to or greater than a predetermined threshold value Lth (Ld-Lc≧Lth). The second target trajectory R2 is repeatedly calculated based on the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc obtained at a predetermined interval. Ld-Lc in pattern 2 is greater than zero (Ld-Lc>0). In pattern 2, the deviation amount Ld of the host vehicle V1 relative to the first target trajectory R1 is greater than the shift amount Lc of the host vehicle V1 relative to the central trajectory MT by at least Lth. In other words, in pattern 2, the deviation amount of the host vehicle V1 from the first target trajectory R1 is large, and the host vehicle V1 is located closer to the central trajectory MT than the first target trajectory R1. As in pattern 2, when the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is equal to or greater than a predetermined threshold value Lth (Ld-Lc≧Lth), the central trajectory MT is calculated as the second target trajectory R2. Also, as in pattern 2, when it is determined that the lateral position (X coordinate value) of the first target trajectory R1 and the lateral position (X coordinate value) of the vehicle V1 are on opposite sides of the center trajectory MT, the processor 10 calculates the center trajectory MT as the second target trajectory R2. Furthermore, as in pattern 2, when it is determined that the lateral position (X coordinate value) of the first target trajectory R1 and the lateral position (X coordinate value) of the host vehicle V1 are on opposite sides of the center trajectory MT, and the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is equal to or greater than a predetermined threshold value Lth (Ld-Lc≧Lth), the processor 10 calculates the center trajectory MT as the second target trajectory R2. If it is determined that the host vehicle V1 is located closer to the central trajectory MT than the first target trajectory R1 during deviation, and / or if the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is greater than or equal to a predetermined threshold value Lth (Ld-Lc≧Lth), the operation of the host vehicle V1 is controlled to follow the central trajectory MT, so that a large steering amount is not required to resolve the deviation state, and the amount of lateral jerk caused by the steering amount can be suppressed. When the processor 10 calculates the first target trajectory R1 by adding an offset amount to the center trajectory MT, the center trajectory MT becomes the reference for the first target trajectory R1. When the host vehicle V1 follows the center trajectory MT as the second target trajectory R2, the coefficient K, which is the degree of reflection of the offset amount, is set to 0 (K=0). When the first target trajectory R1 is calculated by adding an offset amount to the center trajectory MT, the offset amount for the center trajectory MT is canceled by multiplying the offset amount by K=0, and the center trajectory MT is calculated as the second target trajectory R2. In the situation of pattern 2, the host vehicle V1 deviates laterally from the first target trajectory R1 by a deviation amount Ld, sets the center trajectory MT as the second target trajectory R2, and autonomously drives the host vehicle V1 to follow this trajectory. The host vehicle V1 travels along the return trajectory RB and then drives to follow the second target trajectory R2. The second target trajectory R2 is calculated at a predetermined interval, and the second target trajectory R2 calculated every hour gradually approaches the first target trajectory. When the processor 10 determines that the second target trajectory R2 overlaps with the first target trajectory R1 and the host vehicle V1 has returned to the first target trajectory R1, the processor 10 drives the host vehicle V1 along the first target trajectory.

[0026] In this embodiment, the predetermined threshold Lth is a positive value (Lth>0), and the narrower the lane width Lb of the road RT of the host vehicle V1, the lower the value. The road RT of the host vehicle V1 can be identified by referring to the map information 5 based on the current position obtained from the sensor 2 or the host vehicle information acquisition device 3. The map information 5 includes lane information 51 including the lane width Lb of each lane, and the processor 10 acquires the lane width Lb associated with the identification information of the identified road RT from the lane information 51. The lane width Lb may also be calculated as the distance in the X-axis direction between the right road boundary LR and the left road boundary LL calculated based on the detection information of the sensor 2. The lane width Lb may also be calculated based on the distance from the center track MT to the road boundaries LR and LL. By setting the predetermined threshold Lth to a lower value as the road lane width Lb of the road becomes narrower, the movement amount required to return to the first target trajectory R1 can be appropriately adjusted according to the lane width Lb.

[0027] <Pattern 3> Pattern 3 shows an example of the second target trajectory R2 calculated when the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is greater than 0 and less than the predetermined threshold Lth (Lth>Ld-Lc>0). In pattern 3, the deviation amount Ld of the lateral position of the host vehicle V1 relative to the first target trajectory R1 is greater than the shift amount Lc of the host vehicle V1 relative to the center trajectory MT, and the difference is less than the predetermined threshold Lth. When the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc is greater than 0 and less than the predetermined threshold Lth (Lth>Ld-Lc>0) as in pattern 3, a corrected trajectory AT passing through any position between the center trajectory MT and the first target trajectory R1 is calculated as the second target trajectory R2. In the situation of pattern 3, the calculated corrected trajectory AT becomes the second target trajectory R2. The corrected trajectory AT is repeatedly calculated based on the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc calculated at a predetermined interval. The corrected trajectory AT changes from moment to moment depending on the position of the host vehicle V1, and gradually approaches the first target trajectory R1. FIG. 3 shows two situations (a) and (b) according to the relative position of the host vehicle V1 with respect to the first target trajectory R1 and the target trajectory MT. In the example of pattern 3(a), the host vehicle V1 is located between the first target trajectory R1 and the central trajectory MT. When the lateral position of the host vehicle V1 is located between the first target trajectory R1 and the central trajectory MT, the processor 10 calculates a corrected trajectory AT that passes through any position between the central trajectory MT and the first target trajectory R1 as the second target trajectory R2. In the example of pattern 3(a), if the condition Lth>Ld-Lc>0 is satisfied, that is, the difference between the deviation amount Ld and the shift amount Lc is greater than 0 and less than a predetermined threshold value Lth, the processor 10 calculates one of the correction orbits AT that passes through a position between the central orbit MT and the first target orbit R1 as the second target orbit R2. In addition, in the example of pattern 3(a), when the condition Lth>Ld-Lc>0 is satisfied and the lateral position of the vehicle V1 is between the first target trajectory R1 and the central trajectory MT, the processor 10 calculates a corrected trajectory AT that passes through any position between the central trajectory MT and the first target trajectory R1 as the second target trajectory R2. The second target trajectory R2 calculated in the example of pattern 3(a) passes through a position between the central trajectory MT and the first target trajectory R1, but does not include positions on the central trajectory MT or the first target trajectory R1. In the example of pattern 3(a), when multiple correction trajectories AT are calculated, the processor 10 calculates a trajectory from among the multiple correction trajectories AT that passes through a position close to the host vehicle V1 as the second target trajectory R2. The correction trajectory AT that passes through a position closest to the host vehicle V1 may be calculated as the second target trajectory R2. The correction trajectory AT that passes through a position less than a predetermined distance from the host vehicle V1 and that generates the lowest lateral jerk in the host vehicle V1 on the return trajectory RB from the current position of the host vehicle V1 to the correction trajectory AT (second target trajectory R2) may be selected and calculated as the second target trajectory R2. By setting the correction trajectory AT from among the multiple correction trajectories AT that passes through a position close to the host vehicle V1 as the second target trajectory R2, the occurrence and amount of lateral jerk can be suppressed when eliminating the deviation. In this way, when it is determined that the host vehicle V1 is located between the first target trajectory R1 and the central trajectory MT during deviation, and / or when the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is greater than 0 and less than the predetermined threshold Lth (Lth > Ld - Lc > 0), a corrected trajectory AT that passes through a position between the central trajectory MT and the first target trajectory R1 is set as the second target trajectory R2, and the driving of the host vehicle V1 is controlled to follow the second target trajectory R2. This prevents a large steering amount from being required to resolve the deviation state, and the amount of lateral jerk caused by the steering amount can be suppressed. When multiple corrected trajectories AT are calculated, a corrected trajectory AT that passes through a position close to the host vehicle V1 is set as the second target trajectory R2, thereby preventing a large steering amount from being required to resolve the deviation state, and the deviation state can be resolved while suppressing the amount of lateral jerk caused by the steering amount.

[0028] In the example of pattern 3(b), the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on opposite sides of the center trajectory MT. The first target trajectory R1 is on the right or left side of the center trajectory MT, and the host vehicle V1 is on the opposite side. In an example of pattern 3(b), if the processor 10 determines that the difference between the deviation amount Ld and the shift amount Lc is greater than 0 and less than a predetermined threshold value Lth, i.e., Lth>Ld-Lc>0, is satisfied, and that the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on opposite sides of the central trajectory MT, the processor 10 calculates a corrected trajectory AT that passes through any position between the central trajectory MT and the first target trajectory R1 as the second target trajectory R2. This is common to pattern 3(a) in that, if multiple corrected trajectories AT are calculated, the trajectory that passes through a position closest to the host vehicle V1 is calculated as the second target trajectory R2. The above description is incorporated herein to avoid redundancy. In the example of pattern 3(b), the position of the host vehicle V1 is closer to the center trajectory MT than to the target trajectory R1. However, if the center trajectory MT is set as the second target trajectory R2, there is a possibility that the host vehicle V1 will not be able to return to the first target trajectory R1. For this reason, if the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on opposite sides of the center trajectory MT and the condition Lth>Ld-Lc>0 is satisfied, in which the difference between the deviation amount Ld and the shift amount Lc is greater than 0 and less than a predetermined threshold value Lth, a corrected trajectory AT that passes through any position between the center trajectory MT and the first target trajectory R1 is calculated as the second target trajectory R2, rather than setting the center trajectory MT as the second target trajectory R2. In this way, if it is determined that the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on opposite sides of the central trajectory MT during deviation, and / or if the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is greater than 0 and less than the predetermined threshold Lth (Lth > Ld - Lc > 0), a corrected trajectory AT that passes through a position between the central trajectory MT and the first target trajectory R1 is set as the second target trajectory R2, and the driving of the host vehicle V1 is controlled to follow the second target trajectory R2. This eliminates the need for a large steering amount to resolve the deviation, and makes it possible to suppress the amount of lateral jerk caused by the steering amount. When multiple corrected trajectories AT are calculated, by setting the corrected trajectory AT that passes through a position close to the host vehicle V1 as the second target trajectory R2, it is possible to eliminate the deviation while suppressing the amount of lateral jerk caused by the steering amount.

[0029] Similar to Pattern 2, in Patterns 3(a) and (b) as well, the predetermined threshold value Lth is a positive value, and it is set to a lower value as the lane width Lb of the driving lane RT of the host vehicle V1 is narrower. The method for obtaining the predetermined threshold value Lth is common to Pattern 2, so the description thereof is incorporated herein by reference. By setting the predetermined threshold value Lth to a lower value as the lane width Lb of the driving lane is narrower, the amount of movement for returning to the first target trajectory R1 can be appropriately adjusted according to the lane width Lb.

[0030] In the present embodiment, in order to calculate a target trajectory that is easy to adjust, the first target trajectory R1 can be calculated using a method of applying an offset amount to the center trajectory MT. The second target trajectory R2 is calculated by calculating a second offset amount based on the first offset amount of the first target trajectory R1 with respect to the center trajectory MT, and applying this second offset amount to the center trajectory MT. In the present embodiment, the correction trajectory AT is calculated by a method of applying an offset amount to the center trajectory MT. The processor 10 adjusts the first offset amount of the first target trajectory R1 with respect to the center trajectory MT in order to calculate the correction trajectory AT. The processor 10 acquires the first offset amount of the first target trajectory R1 with respect to the center trajectory MT. The processor 10 calculates a coefficient K that defines the degree of application of the first offset amount in order to adjust the first offset amount. The coefficient K in Pattern 3 for calculating the correction trajectory AT is a value greater than zero and less than one (0 < K < 1). Incidentally, in Pattern 1, the coefficient K is 1 (K = 1), and the first offset amount calculated for the first target trajectory R1 is applied. That is, the first target trajectory R1 becomes the second target trajectory R2. In Pattern 2, the coefficient K is zero (K = 0), and the first offset amount with respect to the center trajectory MT for calculating the first target trajectory R1 is canceled. That is, the center trajectory MT becomes the second target trajectory R2. In the situation of pattern 3(a)(b), the host vehicle V1, which has deviated laterally from the first target trajectory R1 by a deviation amount Ld, sets the calculated corrected trajectory AT as the second target trajectory R2 in order to resolve the deviation state. The host vehicle V1 is controlled to drive along the second target trajectory R2. The second target trajectory R2 is calculated every hour based on the difference between the deviation amount Ld and the shift amount Lc, which are calculated at a predetermined interval. The processor 10 sequentially sets the calculated second target trajectory R2 as the target trajectory for the autonomous driving of the host vehicle V1. The host vehicle V1 follows the second target trajectory R2 after traveling along the return trajectory RB. The repeatedly calculated second target trajectory R2 gradually approaches the first target trajectory R1. Thereafter, when the second target trajectory R2 overlaps with the first target trajectory R1, the host vehicle V1 returns to the initially optimized first target trajectory R1.

[0031] Here, a method for calculating the first target trajectory R1 and the second target trajectory R2 using an offset amount with respect to the central trajectory MT will be described with reference to Fig. 4. Fig. 4 shows a block configuration of the processor 10 of the driving assistance device 1 that calculates the second target trajectory. Each function will be described below. (1) Explain the central orbit calculation function. The processor 10 calculates the position of the center track MT using the detection information of the sensor 2. The processor 10 recognizes the positions of the right lane boundary LR and the left lane boundary LL of the lane RT based on the detection information of the sensor 2, and calculates the center position (Xc, Yc) along the road width direction of the lane RT as the center track MT. The processor 10 may refer to the lane information 51 of the lane RT and obtain the center position information on the lane RT included in the lane information 51 as the center track MT. The position (Xc, Yc) of the center track MT is output to the first target trajectory calculation function and the offset calculation function. (2) The first target trajectory calculation function will be explained. The first target trajectory R1 is a trajectory optimized for the driving of the host vehicle traveling along the route to the destination. The processor 10 calculates the first target trajectory R1 (Xdr, Ydr) by adding an offset amount (first offset amount) to the central trajectory MT. This first target trajectory R1 (Xdr, Ydr) is offset from the central trajectory MT by the first offset amount. The first offset amount is output to the calculation function of the coefficient K and the second offset amount calculation function. (3) Explain the calculation function of the coefficient K. When it is determined that the host vehicle V1 has deviated from the first target trajectory R1, the processor 10 adjusts the degree of reflection of the first offset amount with respect to the first target trajectory R1 by the coefficient K, and adds the adjusted second offset amount to the central trajectory MT to calculate the second target trajectory R2. The processor 10 sets an offset coefficient K for adjusting the first offset amount. The offset coefficient K is the degree of reflection of the first offset amount. The offset coefficient K takes a value of 0 ≤ K ≤ 1. The offset coefficient K is set according to the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc calculated based on the current position of the input host vehicle. That is, the offset coefficient K is set according to the latest positional relationship among the current position of the deviating host vehicle V1, the first target trajectory R1, and the central trajectory MT. The coefficient K takes a value between 0 and 1 (0 ≤ K ≤ 1) with respect to the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc, as shown in the graph of the coefficient K calculation function in FIG. 4. When the difference (Ld - Lc) is zero, the coefficient K is 1, and when the difference (Ld - Lc) is the threshold value Lth, the coefficient K is 0. When the difference (Ld - Lc) is greater than 0 and less than the threshold value Lth, the coefficient K is 0 < K < 1. As the difference (Ld - Lc) increases, K takes a small value. The calculated coefficient k is output to the second offset amount calculation function. (4) Explain the second offset amount calculation function. The processor 10 calculates the second offset amount by multiplying the first offset amount by the coefficient K. The calculated second offset amount is output to the offset calculation function. (5) Explain the offset calculation function. The processor 10 calculates the second target trajectory R2 by applying a second offset amount to the central trajectory MT (Xc, Yc). FIG. 4 includes a diagram illustrating the offset calculation method. The central trajectory MT is divided at points P1 to P5. Second offset amounts d1 to d5 are applied to each of the points dividing the trajectory. The second offset amounts d1 to d5 are obtained by multiplying the first offset amount by an offset coefficient K. Each of the points P1 to P5 dividing the central trajectory MT is offset by the specified offset amount d1 to d5 to obtain each of the offset points P1' to P5', and these are then connected to obtain the second target trajectory R2. The first target trajectory R1 is also obtained by applying the first offset amount to the central trajectory MT. The calculated second target trajectory R2 is output to a second target trajectory output function. (6) The second target trajectory output function will be explained. Each time the second target trajectory R2 is calculated, the processor 10 outputs the newly calculated second target trajectory R2 to the vehicle controller 200. The processor 10 sequentially outputs the calculated second target trajectory R2 to the vehicle controller 200 as the target trajectory of the host vehicle V1 from when the host vehicle V1 deviates from the first target trajectory R1 until the host vehicle V1 returns to the first target trajectory R1. The vehicle controller 200 causes the host vehicle V1 to autonomously drive so as to follow the set second target trajectory R2.

[0032] In this embodiment, the first target trajectory R1 can be calculated by applying a first offset amount to the central trajectory MT. Furthermore, the first offset amount defining the first target trajectory R1 is used to calculate a second offset amount corresponding to the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc, and the second target trajectory R2 can be calculated by applying the second offset amount to the central trajectory MT. In this manner, the second target trajectory R2 in which lateral G jerk is suppressed can be calculated while reducing the computational load. Furthermore, an offset coefficient K can be set and the second offset amount can be calculated by multiplying the first offset amount by the coefficient K. Therefore, the second target trajectory R2 with high tracking ability can be calculated while reducing the occurrence of lateral G jerk.

[0033] The processor 10 calculates an offset coefficient K based on the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc. As shown in FIG. 3, in pattern 1 where 0 ≧ Ld - Lc, the offset coefficient K = 1, and the second target trajectory R2 becomes the first target trajectory R1. In pattern 2 where Ld - Lc ≧ Lth, the offset coefficient K = 0, and the second target trajectory R2 becomes the center trajectory MT. In pattern 3 where Lth > Ld - Lc > 0, the offset coefficient 0 < K < 1, and the second target trajectory R2 becomes one of the one or more correction trajectories AT. Since the offset amount can be changed according to the first target trajectory R1, the center trajectory MT, and the current position of the host vehicle V1, it is possible to calculate a highly following second target trajectory R2 while reducing the occurrence of lateral G jerk.

[0034] Although not particularly limited, the processor 10 calculates an offset coefficient based on the following formula (1).

Equation

[0035] When the longitudinal distance along the traveling direction between the current position of the host vehicle V1 and the point defining the first target trajectory R1 is long, the processor 10 sets the offset coefficient K to a smaller value than when the longitudinal distance is short. The point defining the second target trajectory R2 is a point defining multiple sections included in the route (traveling path RT). For example, P1' to P4' shown on the target trajectory in the offset calculation diagram of FIG. 4 are points defining the second target trajectory R2. The traveling direction is the direction along the extension direction of the route (the traveling direction of the vehicle). The longitudinal distance is the distance between the current position of the host vehicle V1 and the point defining the first target trajectory R1 that is closest along the traveling direction from the current position of the host vehicle V1. When the longitudinal distance along the traveling direction between the current position of the host vehicle V1 and the point defining the first target trajectory R1 is long, the distance traveled by the host vehicle V1 to reach the forward point becomes relatively long, and therefore the change in the lateral position can be set to a relatively small value. By setting the offset coefficient of the second target trajectory R2 relatively small, it is possible to reduce the change in steering amount and the occurrence of lateral G jerk in the section with a long vertical distance along the direction of travel between the current position of the vehicle V1 and the point defining the first target trajectory R1, and to calculate a second target trajectory R2 with high tracking performance.

[0036] FIG. 5 is a flowchart showing the procedure of the calculation process of the second target trajectory described above. 5 shows a subroutine of the trajectory calculation method (S16) when the deviation amount Ld is greater than the shift amount Lc in S14 of FIG. 2. The calculation process of the second target trajectory R2 described below is executed when it is determined that the host vehicle V1 has deviated from the first target trajectory R1 (YES in S7 of FIG. 2). In this processing example, the calculation process of the first target trajectory R1 in S3 of FIG. 2 is calculated by adding an offset amount to the central trajectory MT. The processes of S1-S15 and S20-S22 of FIG. 2 are also used herein.

[0037] Each process shown in FIG. 5 will be explained below. When the calculation process for the second target trajectory R2 is started (S31), the processor 10 acquires the current position of the host vehicle V1 at a predetermined period and calculates, hourly, a deviation amount Ld of the lateral position of the host vehicle V1 relative to the first target trajectory R1 based on the acquired current position (S32). The processor 10 acquires the current position of the host vehicle V1 at a predetermined period and calculates, hourly, a shift amount Lc of the lateral position of the host vehicle V1 relative to the center trajectory MT based on the acquired current position (S33). The processor 10 calculates the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc at a predetermined period (S34). If the deviation amount Ld is greater than the shift amount Lc (0≧Ld−Lc) (YES in S35), the first target trajectory R1 and the lateral position of the host vehicle V1 are on the same side of the center trajectory MT. In this case, the first target trajectory R1 is closer to the host vehicle V1 than the center trajectory MT. If the deviation amount Ld is greater than the shift amount Lc (0≧Ld−Lc) (YES in S35), the offset coefficient K is set to 1 (K=1) as in pattern 1 of FIG. 3 (S36). If the offset coefficient K is 1, the first target trajectory R1 is selected. The processor 10 calculates a second target trajectory R2 that overlaps with the first target trajectory R1 (S37). Once the second target trajectory R2 is calculated, the processes of S20-S22 are performed (S38).

[0038] When the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is less than zero (0≧Ld - Lc) (NO in S35), the processor 10 determines that the difference (Ld - Lc) is greater than zero (0 < Ld - Lc) (S40). When Ld - Lc is greater than or equal to a predetermined threshold value Lth (Ld - Lc≧Lth) (YES in S41), the processor 10 assumes a scenario where the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on the opposite side of the center trajectory MT (S42). When the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is greater than or equal to the predetermined threshold value Lth (Ld - Lc≧Lth), the offset coefficient K is set to 0 (K = 0) as in Pattern 2 of FIG. 2 (S43). When the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is greater than or equal to the predetermined threshold value Lth, the second target trajectory R2 overlapping the center trajectory MT is calculated (S44). When the second target trajectory R2 is calculated, the processes of S20 - S22 are performed (S38). Specifically, the processor 10 drives the host vehicle V1 along the second target trajectory R2 (S20), and updates the second target trajectory R2 calculated at a predetermined cycle as the target trajectory every hour until the second target trajectory R2 overlaps the first target trajectory R1. When the host vehicle V1 returns to the first target trajectory R1 (YES in S21), the host vehicle is driven along the first target trajectory R1 (S22). Note that Lth used in S41 can be set according to the lane width Lb. When the processor 10 determines that the difference (Ld - Lc) is greater than zero (0 < Ld - Lc) (S40), the lane width Lb is calculated based on the pair of running lane boundaries LL, LR (S51), and the predetermined threshold value Lth is calculated based on the lane width Lb (S52). The predetermined threshold value Lth is set to a lower positive value as the lane width Lb of the running lane RT of the host vehicle V1 is narrower. The processes of S51 and S52 are additional processes and are executed as necessary.

[0039] If the difference (Ld - Lc) between the deviation amount Ld and the shift amount Lc is not equal to or greater than the predetermined threshold value Lth (Ld - Lc ≥ Lth) (NO in S41), it is determined that Lth > Ld - Lc > 0 (S45). The following situations are assumed: the lateral position of the host vehicle V1 is between the first target trajectory R1 and the center trajectory MT; or the lateral position of the first target trajectory R1 and the lateral position of the host vehicle V1 are on opposite sides of the center trajectory MT (S46). The processor 10 refers to equation (1) and uses this to calculate the offset coefficient K (S48). Using the offset coefficient K, the processor 10 selects one of the correction trajectories AT that passes through any lateral position (position along the road width direction) between the center trajectory MT and the first target trajectory RT, and calculates the selected correction trajectory AT as the second target trajectory R2 (S49). In this process, the correction trajectory AT does not include the center trajectory MT. The corrected trajectory AT is calculated every hour based on the difference (Ld-Lc) between the deviation amount Ld and the shift amount Lc, which are successively detected at a predetermined cycle from the timing when the deviation is determined. The processor 10 calculates one corrected trajectory AT as the second target trajectory R2 (S50).

[0040] Once the second target trajectory R2 is calculated, the process proceeds to S38, where steps S20-S22 are performed. The processor 10 executes driving control so that the host vehicle V1 follows the second target trajectory R2 (S20). The second target trajectory R2 is calculated hourly based on the difference between the deviation amount Ld and the shift amount Lc, which are calculated at a predetermined interval. The processor 10 sequentially sets the calculated second target trajectory R2 as the target trajectory for the host vehicle's autonomous driving. The host vehicle V1 travels along the return trajectory RB and then follows the second target trajectory R1. The second target trajectory R2 gradually approaches the first target trajectory R1. If the processor 10 determines that the second target trajectory R2 overlaps with the first target trajectory R1 (they share a common trajectory) and that the host vehicle V1 has returned to the first target trajectory R1 (YES in S21), the processor 10 drives the host vehicle along the first target trajectory R1 (S22). When the second target trajectory R2 and the first target trajectory R1 overlap, or when the difference in lateral distance between the second target trajectory R2 and the first target trajectory R1 becomes less than a predetermined threshold, the processor 10 determines that the vehicle V1 has returned to the first target trajectory R1, and performs autonomous driving of the vehicle V1 so that it follows the first target trajectory R1. [Explanation of symbols]

[0041] 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, 21... camera, 22... radar device, 3... vehicle information acquisition device, 4... road recognition 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 for use in a processor, which causes a vehicle to autonomously drive so that the vehicle travels along a target trajectory, comprising: The processor: Acquire vehicle information including the current location of the vehicle; calculating a first target trajectory along which the host vehicle is to travel; calculating a deviation amount of a lateral position of the host vehicle from the first target trajectory based on the host vehicle information; When it is determined that the host vehicle has deviated from the first target trajectory based on the deviation amount, Acquire road information about the road on which the vehicle is traveling; Calculating a center trajectory that passes through the center of the lane boundary determined based on the lane information; calculating a shift amount of a lateral position of the vehicle relative to the center track based on the vehicle information; If the deviation amount is greater than the shift amount, calculating a difference between the deviation amount and the shift amount; calculating, according to the difference, either the central orbit or a corrected orbit that passes through a position between the central orbit and the first target orbit, as a second target orbit; A driving control method for driving the vehicle along the second target trajectory.

2. 2. The driving control method according to claim 1, wherein the processor calculates the center track as the second target track when it determines that the lateral position of the first target track and the lateral position of the vehicle are on opposite sides of the center track.

3. The driving control method according to claim 1 , wherein the processor calculates the center trajectory as the second target trajectory when the difference between the deviation amount and the shift amount is equal to or greater than a predetermined threshold.

4. The driving control method according to claim 3 , wherein the processor sets the predetermined threshold to a positive value, and the smaller the lane width of the road on which the host vehicle is traveling, the lower the value.

5. 2. The driving control method according to claim 1, wherein, when the lateral position of the host vehicle is between the first target trajectory and the central trajectory, the processor calculates, as the second target trajectory, the corrected trajectory that passes through a position between the central trajectory and the first target trajectory.

6. 2. The operation control method according to claim 1, wherein, when the difference between the deviation amount and the shift amount is greater than 0 and less than a predetermined threshold, the processor calculates the corrected trajectory that passes between the central trajectory and the first target trajectory as the second target trajectory.

7. The driving control method according to claim 6 , wherein the processor sets the predetermined threshold to a positive value, and the narrower the lane width of the road on which the host vehicle is traveling, the lower the value.

8. The operation control method according to any one of claims 1 to 7, wherein the processor calculates the first target trajectory by adding an offset amount to the central trajectory.

9. The operation control method according to claim 8 , wherein the processor sets an offset coefficient for adjusting the offset amount.

10. The driving control method according to claim 9 , wherein the processor calculates the offset coefficient based on the difference between the deviation amount and the shift amount.

11. The operation control method according to claim 9 , wherein the processor calculates the offset coefficient based on the following equation (1): [Equation 2]

12. 10. The driving control method according to claim 9, wherein the processor sets the offset coefficient to a smaller value when the longitudinal distance between the current position of the vehicle and the point defining the first target trajectory is long than when the longitudinal distance is short.

13. The driving control method according to claim 1 , wherein the processor calculates the second target trajectory that overlaps with the first target trajectory when the difference between the deviation amount and the shift amount is equal to or less than 0.

14. A driving control device that includes a processor and autonomously drives a vehicle so that the vehicle travels along a target trajectory, The processor: Acquire vehicle information including the current location of the vehicle; calculating a first target trajectory along which the host vehicle is to travel; calculating a deviation amount of a lateral position of the host vehicle from the first target trajectory based on the host vehicle information; When it is determined that the host vehicle has deviated from the first target trajectory based on the deviation amount, Acquire road information about the road on which the vehicle is traveling; Calculating a center trajectory that passes through the center of the lane boundary determined based on the lane information; calculating a shift amount of a lateral position of the vehicle relative to the center track based on the vehicle information; If the deviation amount is greater than the shift amount, calculating a difference between the deviation amount and the shift amount; calculating, according to the difference, either the central orbit or a corrected orbit that passes through a position between the central orbit and the first target orbit, as a second target orbit; a driving control device that drives the vehicle along the second target trajectory;

Images