JP2019197399A - Route determination device of vehicle - Google Patents

Abstract

To provide a route determination device for a vehicle that selects and avoids a travelable area, even when the travelable area is not correctly recognized, regardless there is the travelable area that avoids an obstacle ahead of the vehicle.SOLUTION: The present invention relates to a route determination device that determines a travel route of a vehicle, and includes an obstacle detection portion, a travelable area detection portion, a preceding vehicle lane deviation detection portion, and a route generation portion. The obstacle detection portion detects an obstacle ahead of an own lane. The travelable area detection portion detects a travelable area outside the own lane. The preceding vehicle lane deviation detection portion detects that the preceding vehicle in front of the vehicle has traveled to an area outside the own lane. When the obstacle detection portion avoids the obstacle detected by the obstacle detection portion, the route generation portion generates a travel plan in which the vehicle travels in the travelable area when the travelable area is detected and travels on a travel route following the preceding vehicle when the travelable area is not detected and it is detected that the preceding vehicle travels in the area outside the own lane of the vehicle.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a route determination device that is mounted on a vehicle capable of automatically performing part or all of the driving of the vehicle and determines a travel route of the vehicle.

  For example, Patent Literature 1 describes a driving support device that supports driving of a vehicle. According to Patent Document 1, this driving support device detects an obstacle around the host vehicle and determines that there is a possibility of colliding with the obstacle. Is configured to select an obstacle-free route and correct the traveling direction of the host vehicle to the obstacle-free route. Further, when there is no unobstructed route, the route of the direction of the low-impact obstacle with the least seriousness of the accident at the time of collision is selected, and the traveling direction of the own vehicle is corrected to the selected route. ing.

JP 2017-117191 A

  By the way, an obstacle in front of the host vehicle, a free space where the vehicle can travel, and the like are recognized based on, for example, analysis of an image acquired by a camera mounted on the vehicle and output of a sensor or the like. However, particularly when free space is detected by a sensor or the like, sufficient detection accuracy may not be ensured. And when detection accuracy is low and the free space which can drive | work is not detected correctly, even if an obstacle can be detected, the optimal avoidance operation | movement cannot be taken.

  In this regard, Patent Document 1 also detects an obstacle or a no-failure route only by analyzing a camera image, and the no-failure route may not be detected correctly. In this case, the selected avoidance operation is not necessarily optimal. Therefore, it is desired to develop an apparatus configured to be able to take a more appropriate avoidance operation even when the free space is not correctly detected.

  In order to solve the above problems, the present invention does not have a region that can be traveled from image analysis or the like even though there is a region that can travel to avoid an obstacle ahead of the host vehicle. In such a case, an improved vehicle route determination device is provided so that avoidance traveling can be performed more appropriately.

  The present invention is a route determination device that determines a travel route of a vehicle, and is mounted on a vehicle capable of automatically performing all or part of the driving of the vehicle. The route determination device includes an obstacle detection unit, a travelable region detection unit, a preceding lane departure detection unit, and a route generation unit.

  The obstacle detection unit detects an obstacle in front of the own lane which is a traveling lane on which the vehicle travels. The travelable area detection unit detects a travelable area outside the own lane. This travelable area includes, for example, lanes other than the own lane, road shoulders, and the like. The preceding lane departure detection unit detects that the preceding vehicle in front of the vehicle traveling in the own lane has traveled to an area outside the own lane.

  The route generation unit generates a travel route that avoids the obstacle when the obstacle detection unit detects the obstacle. Here, the route generation unit is configured to generate a travel route for traveling in the detected travelable region when the travelable region is detected by the travelable region detection unit. Further, the route generation unit is a case where the travelable region detection unit does not detect a travelable region, and the preceding lane departure detection unit detects that the preceding vehicle travels to a region outside the own lane. Is configured to generate a travel route that follows the preceding vehicle.

  According to the present invention, even when an obstacle is detected in front of the vehicle and it is determined that there is no travelable area outside the own lane in which the vehicle travels, the preceding vehicle moves to the area outside the own lane. When traveling, a travel route is generated based on the behavior of the preceding vehicle. Therefore, when there is a travelable area that cannot be acquired from camera images, sensors, etc., by following the preceding vehicle, the travelable area can be traveled, and obstacles can be avoided by an appropriate route. Can do.

It is a block diagram which shows the structure of the automatic driving device to which the route determination apparatus of Embodiment 1 of this invention was applied. It is the figure which represented the production | generation routine of the obstacle avoidance path | route by the path | route production | generation part of Embodiment 1 of this invention with the flowchart. It is a block diagram which shows the structure of the automatic driving device to which the route determination apparatus of Embodiment 2 of this invention was applied. It is the figure which represented the production | generation routine of the obstacle avoidance path | route by the path | route production | generation part of Embodiment 2 of this invention with the flowchart.

  Embodiments of the present invention will be described below with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and the description thereof is simplified or omitted.

Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a configuration of an automatic driving device to which the route determination device of the present embodiment is applied. The automatic driving device is mounted on a host vehicle such as a passenger car. The automatic driving device executes automatic driving of the host vehicle. The automatic driving means that driving operations such as acceleration, deceleration, and steering of the host vehicle are executed without depending on the driving operation of the driver of the host vehicle. In the automatic driving apparatus according to the present embodiment, a plurality of automatic driving modes having different parameter settings related to driving operation during automatic driving are defined. The automatic driving device according to the present embodiment performs automatic driving of the host vehicle in a mode selected from a plurality of automatic driving modes, and changes the automatic driving being performed to manual driving according to the traveling conditions of the host vehicle. Switch.

  In this embodiment, the case where the route determination device is applied to an automatic driving device will be described. However, the route determination device is applied to other driving support devices that can automatically perform part of the driving of the vehicle. can do. For example, the driving support device is configured to automatically perform only the obstacle avoidance operation when the obstacle described below is detected, regardless of the driver's intention, according to the plan generated by the route determination device. It may be what was done.

  As shown in FIG. 1, the automatic driving apparatus includes an external sensor 2, a GPS (Global Positioning System) receiving unit 4, an internal sensor 6, a map database 8, and an ECU 10. Although not shown, the navigation system, the actuator, the HMI (Human Machine Interface), and auxiliary equipment are provided.

  The external sensor 2 is a detection device that detects an external situation that is surrounding information of the host vehicle. The external sensor 2 includes at least one of a camera, a radar (Radar), and a rider (LIDER: Laser Imaging Detection and Ranging).

  The camera is an imaging device that captures an external situation of the host vehicle V, and outputs imaging information regarding the external situation of the host vehicle to the ECU 10. The radar detects obstacles outside the host vehicle using radio waves. The rider uses light to detect an obstacle outside the host vehicle. For example, the radar and the rider can output the distance or direction to the obstacle to the ECU 10 as obstacle information regarding the obstacle. The cameras, riders, and radars do not necessarily have to be provided in duplicate.

  The GPS receiving unit 4 receives signals from three or more GPS satellites, acquires position information indicating the position of the host vehicle V, and outputs the position information to the ECU 10. The internal sensor 6 is a detector that detects information according to the traveling state of the host vehicle, the steering operation amount of the steering operation by the driver of the host vehicle, and information regarding the state of the driver of the host vehicle V. The map database 8 is a database provided with map information. The map database 8 is formed, for example, in an HDD (Hard disk drive) mounted on the host vehicle V. The map information includes, for example, information on the type of road such as an expressway or a general road, road position information, road shape information, and intersection and branch point position information. The road shape information includes, for example, a curve, a straight line type, a curve curvature, and the like. The map database 8 may be stored in a computer of a facility such as an information processing center that can communicate with the host vehicle V.

  The ECU 10 is connected to a plurality of actuators that execute traveling control of the host vehicle. The actuator includes, for example, a throttle actuator, a brake actuator, and a steering actuator. The ECU 10 controls the driving force of the host vehicle by a control signal to the throttle actuator. In the case where the host vehicle V is a hybrid vehicle or an electric vehicle, a control signal from the ECU 10 is input to a motor as a power source and the driving force is controlled without including a throttle actuator. Moreover, ECU10 controls the braking force given to the wheel of the own vehicle V by the control signal to a brake actuator. Further, the ECU 10 controls the steering torque of the host vehicle V by a control signal to the steering actuator.

  The ECU 10 controls the driving of the host vehicle. The ECU 10 is an electronic control unit having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. However, the ECU 10 may be composed of a plurality of electronic control units. In the ECU 10, various functions are realized by loading a program stored in the ROM into the RAM and executing it by the CPU. The functions realized by the ECU 10 include a function as a route determination device according to the present invention.

  In the present embodiment, when the ECU 10 functions as a route determination device, the ECU 10 detects the object detection unit 12, the white line detection unit 14, the free space detection unit 16, the preceding vehicle tracking unit 18, the obstacle detection unit 20, and the preceding lane. A departure detection unit 22 and a route generation unit 24 are included. However, these do not exist as individual hardware, but are virtually realized when a program stored in the memory is executed.

  The object detection unit 12 is configured to detect a vehicle, a motorcycle, and a vehicle in front of the own lane that is the travel route of the own vehicle from image information analyzed based on a rule base or a deep learning base or information from sensors such as a radar and a rider. Detect obstacles such as cardboard.

  The white line detection unit 14 detects a white line from image information analyzed on a rule basis or a deep learning basis.

  The free space detection unit 16 receives input image information or information from a sensor such as a radar or a rider from image information analyzed based on a rule base or a deep learning base, or from information from a sensor such as a radar or a rider. A free space that is an area in which the host vehicle can travel by changing the course is detected. The free space includes lanes and shoulders outside the own lane. The free space detection unit 16 corresponds to the “runnable region detection unit” in the present invention.

  When the object detection unit 12 detects a preceding vehicle that is a vehicle immediately preceding the host vehicle traveling in the own lane, information on the preceding vehicle is input to the preceding vehicle tracking unit 18. The preceding vehicle tracking unit 18 tracks the preceding vehicle for each image frame based on the image information of the preceding vehicle and the map database, and detects the travel route on the map of the preceding vehicle.

  The obstacle detection unit 20 receives information on the obstacle detected by the object detection unit 12, information on the white line detected by the white line detection unit 14, and information on the free space detected by the free space detection unit. Is done. Based on these pieces of information, the obstacle detection unit 20 detects an obstacle present in the own lane, the adjacent lane, or the free space where the host vehicle travels.

  The preceding vehicle lane departure detection unit 22 receives the travel route of the preceding vehicle detected by the preceding vehicle tracking unit 18 and the white line information detected by the white line detection unit 14. Based on this input information, the preceding lane departure detecting unit 22 detects the lane departure of the preceding vehicle, that is, that the preceding vehicle has traveled to an area outside the own lane.

  The route generation unit 24 includes information on the white line detected by the white line detection unit 14, information on the free space detected by the free space detection unit 16, information on the travel route of the preceding vehicle detected by the preceding vehicle tracking unit 18, Information on the obstacle detected by the obstacle detection unit 20 and information on the lane departure of the preceding vehicle detected by the preceding lane departure detection unit 22 are input. The route generation unit 24 generates a route for avoiding the obstacle based on the input information.

  FIG. 2 is a flowchart representing a routine for generating an obstacle avoidance route by the route generation unit 24. In step S102 of FIG. 2, it is determined whether there is an obstacle ahead of the own lane. This determination is performed based on the information on the obstacle detected by the obstacle detection unit 20.

  If it is determined in step S102 that there is no obstacle, the process proceeds to step S104, and a lane keeping plan, which is a travel route for traveling while maintaining the current lane during traveling, is generated.

  On the other hand, if it is determined in step S102 that there is an obstacle, the process proceeds to step S106, where a lane change (hereinafter also referred to as “LC”. “LC” is an abbreviation for lane change) is possible. It is determined whether or not there is. In the determination in step S106, it is determined whether there is an obstacle in the adjacent lane based on the information from the obstacle detection unit 20, and it is determined that there is no obstacle and that LC is possible. In this case, it is determined that LC is possible.

  If it is determined in step S106 that LC is possible, the process proceeds to step S108, and an LC plan that is a travel route for changing the lane to an adjacent lane where the vehicle can travel is generated.

  On the other hand, if it is determined in step S106 that LC is not possible, the process proceeds to step S110, and it is determined whether there is a free space (FS) outside the roadway. That is, it is determined whether or not there is an area where the vehicle can travel on the road shoulder. This determination is performed based on information from the free space detection unit 16 and information from the obstacle detection unit 20.

  If it is determined in step S110 that there is a free space, the process proceeds to step S112, and a road shoulder FS travel plan, which is a travel route for traveling in the free space on the road shoulder, is generated.

  On the other hand, if it is determined in step S110 that there is no free space on the road shoulder, the process proceeds to step S114, where it is determined whether the preceding vehicle has deviated from the own lane. This determination is made based on information from the preceding lane departure detection unit 22.

  If it is determined in step S114 that the preceding vehicle has not deviated from the lane, the process proceeds to step S116, and a stop plan on the lane that stops on the own lane before colliding with an obstacle is generated. .

  On the other hand, if it is determined in step S114 that the preceding vehicle has deviated from the lane, the process proceeds to step S118, and a preceding vehicle following plan is generated. The preceding vehicle following plan is a traveling route that follows the preceding vehicle, and is generated based on the traveling route information of the preceding vehicle from the preceding vehicle tracking unit 18. Thereafter, the current process ends.

  The ECU 10 automatically controls the travel of the host vehicle according to the travel plan determined by the above route determination device. The ECU 10 outputs a control signal corresponding to the travel plan to each actuator. Thereby, ECU10 controls driving | running | working of the own vehicle so that the automatic driving | operation of the own vehicle is performed according to a runway and speed plan.

  As described above, according to the present embodiment, even when it is determined that there is an obstacle in front of the own lane, and avoidance traveling to free space cannot be performed from the image or sensor output, When the preceding vehicle is traveling out of the lane, a travel plan that follows the preceding vehicle is formed. Therefore, even if there is a travel route that can avoid an obstacle, an appropriate avoidance travel route is not selected due to low accuracy of image recognition or the like, and a situation where the vehicle stops in its own lane is prevented. Can do.

Embodiment 2. FIG.
FIG. 3 is a block diagram showing a configuration of an automatic driving device to which the route determination device of the present embodiment is applied. In the automatic driving device of FIG. 3, the route determination device, which is one function of the ECU 10, further includes a preceding vehicle tracking unit 30 and a preceding lane departure detection unit 32, and an obstacle avoidance route in the route generation unit 34. The configuration is the same as that of the automatic driving apparatus of FIG.

  The preceding vehicle tracking unit 30 detects the traveling route of the preceding vehicle by the same method as the preceding vehicle tracking unit 18. That is, when the object detection unit 12 detects a preceding vehicle ahead of a preceding vehicle traveling in front of the host vehicle in the own lane, information on the preceding vehicle is input to the preceding vehicle tracking unit 30. The predecessor vehicle tracking unit 30 tracks the predecessor vehicle for each image frame based on the input image information of the predecessor vehicle and the map database, and detects a travel route on the map of the predecessor vehicle.

  Similarly to the preceding lane departure detection unit 22, the preceding lane departure detection unit 32 detects the lane departure of the preceding vehicle lane departure. In other words, the preceding vehicle lane departure detection unit 32 receives the preceding vehicle travel route generated by the preceding vehicle tracking unit 30 and the white line information detected by the white line detection unit 14. Based on this input information, the preceding lane departure detecting unit 32 detects the deviation of the traveling lane on the preceding vehicle traveling route.

  The route generation unit 34 includes information on the white line detected by the white line detection unit 14, information on the free space detected by the free space detection unit 16, information on the travel route of the preceding vehicle detected by the preceding vehicle tracking unit 18, In addition to the information of the obstacle detected by the obstacle detection unit 20 and the information of the lane departure of the preceding vehicle detected by the preceding lane departure detecting unit 22, the preceding vehicle detected by the preceding vehicle tracking unit 30 is further added. And the information on the lane departure of the preceding vehicle detected by the preceding lane departure detecting unit 32 are input. The route generation unit 34 generates a travel route for obstacle avoidance based on the input information.

  FIG. 4 is a flowchart representing a routine for generating an obstacle avoidance route by the route generation unit 34. The routine in FIG. 4 is the same as the routine in FIG. 2 except that the processing in steps S202 and S204 is performed after step S102.

  Specifically, in the routine of FIG. 4, when it is determined in step S102 that there is no obstacle, the process proceeds to step S202, where it is determined whether or not the preceding vehicle has deviated from the lane. This determination is made based on information from the preceding lane departure detection unit 22.

  If it is determined in step S202 that the preceding vehicle has deviated, the process proceeds to step S204, and it is determined whether or not the preceding vehicle has deviated from the lane. This determination is made based on information from the lane departure detection unit 32 in advance.

  If it is determined in step S202 that the preceding vehicle has deviated from the lane, the process proceeds to step S118, and a preceding vehicle following plan is generated. That is, in the control of the present embodiment, even if it is determined that there is no obstacle, if the preceding vehicle and the preceding vehicle are both deviating from the lane, the preceding vehicle following plan is generated.

  On the other hand, if it is determined in step S202 that the preceding vehicle has not deviated from the lane, or if it is determined in step S204 that the preceding vehicle has not deviated from the lane, the process proceeds to step S104, and a lane keep plan is generated. The That is, even when it is determined that there is no obstacle, the maintenance of the own lane is permitted only when the preceding vehicle or the preceding vehicle has not deviated from the lane.

  There are almost unlimited obstacles on the lane, and even if object detection using deep learning or free space detection is performed, there are cases where the obstacle cannot be detected. In the present embodiment, even when no obstacle is detected, the travel route is set so as to follow the preceding vehicle if the preceding vehicle and the preceding vehicle deviate from the lane. Thereby, even when an obstacle cannot be detected, the obstacle can be avoided and collision with the obstacle can be avoided more reliably.

  In the above embodiment, when the number of each element, number, quantity, range, etc. is mentioned, it is mentioned unless otherwise specified or clearly specified in principle. The invention is not limited to the numbers. Further, the structure and the like described in this embodiment are not necessarily essential to the present invention unless otherwise specified or clearly specified in principle.

DESCRIPTION OF SYMBOLS 2 External sensor 4 GPS receiving part 6 Internal sensor 8 Map database 12 Object detection part 14 White line detection part 16 Free space detection part 18 Leading vehicle tracking part 20 Obstacle detection part 22 Leading lane deviation detection part 24 Route generation part 30 Predecessor vehicle Tracking unit 32 Predecessor lane departure detection unit 34 Route generation unit

Claims (1)

A route determination device that is mounted on a vehicle capable of automatic driving and determines a travel route of the vehicle,
An obstacle detection unit for detecting an obstacle in front of the own lane which is a traveling lane on which the vehicle travels;
A travelable area detection unit for detecting a travelable area outside the own lane;
A preceding lane departure detecting unit for detecting that a preceding vehicle in front of the vehicle traveling in the own lane has traveled to an area outside the own lane;
A route generation unit that generates a travel route to avoid the obstacle when an obstacle is detected by the obstacle detection unit;
With
The route generation unit
When a travelable region is detected by the travelable region detection unit, a travel route that travels in the travelable region is generated,
When the travelable area detection unit does not detect a travelable area and the preceding lane departure detection unit detects that the preceding vehicle travels to an area outside the own lane, Generating a travel route that follows the preceding vehicle;
A vehicle route determination device configured as described above.

Images