JP6930368B2 - Vehicle control unit - Google Patents

Description

The present invention relates to a vehicle control device having a function of detecting a traveling lane of the own vehicle.

Patent Document 1 describes a technique for detecting a break in the lane in which the own vehicle is traveling. Specifically, in the lane division detection method described in Patent Document 1, an image of the vicinity of the own vehicle is obtained from each of the side in-vehicle camera installed on the side of the vehicle and the front in-vehicle camera installed in front of the vehicle. get. Then, the lane division existing on the side of the own vehicle is detected from the image taken by the side in-vehicle camera, and the lane division existing in front of the own vehicle is detected from the image taken by the front in-vehicle camera. Further, the position of the lane division in front of the own vehicle is corrected based on the time difference in the shooting timing between the side in-vehicle camera and the front in-vehicle camera and the traveling speed. Then, the corrected lane dividing position in front of the own vehicle is collated with the lane dividing position on the side of the own vehicle, and when the two match, the detection result of the lane dividing position on the side is obtained. Output as a lane detection result.

Japanese Unexamined Patent Publication No. 2016-24777

The image recognition accuracy of the front in-vehicle camera may be significantly reduced due to environmental factors such as backlight while the vehicle is running. In such a case, the lane detection method using the image of the front vehicle-mounted camera described in Patent Document 1 cannot detect the lane. Then, in a situation where the lane cannot be detected, it becomes difficult to continue the automatic driving of the vehicle. When it becomes difficult to drive by automatic driving, it is necessary to promptly prompt the driver to change driving and switch to manual driving. However, in the automatic driving state, it is expected that the driver's awareness of the driving operation is lowered, and it is considered that a certain amount of time is required for the handover to switch from the automatic driving to the manual driving. Therefore, it is desired to develop a means capable of continuing automatic driving in the traveling lane for a longer period of time by automatic driving even when the image recognition accuracy by the front in-vehicle camera is lowered.

Regarding this, when the front in-vehicle camera is affected by the backlight and the image recognition accuracy is lowered, it is conceivable to maintain the automatic driving by detecting the lane using only one of the side in-vehicle cameras. However, the lane recognition accuracy is low only by the side in-vehicle camera, and it is not sufficient as a method for maintaining a relatively long time until the driver changes driving.

An object of the present invention is to solve the above problems, and to solve the above problems, the vehicle control is improved so that the automatic driving can be reliably maintained in the traveling lane even when the image recognition by the front camera is difficult during the automatic driving of the vehicle. It provides a device.

The vehicle control device of the present invention is configured as follows for the purpose of solving the above problems. That is, the vehicle control device includes a front camera that acquires an image of the front of the vehicle and side cameras that acquire images of the left and right sides of the vehicle, and analyzes the images taken by the front camera and the side cameras to determine the traveling lane. It is configured so that it can be detected and the relative position of the vehicle in the detected traveling lane can be calculated. In addition, when a decrease in recognition accuracy of the camera mounted in front of the vehicle is detected, the image acquisition of the front camera is stopped, only the image taken by the side camera is analyzed, and the relative of the vehicle in the traveling lane is analyzed. The position is detected, the surrounding road structure and the distance to the road structure are detected by a laser radar mounted on the vehicle, and the position of the vehicle is calculated by collating with the map information. Then, the relative position of the vehicle calculated by analyzing the image of the side camera is corrected according to the position of the vehicle calculated by using the laser radar.

When the recognition accuracy by the front camera deteriorates, the surrounding structures are identified by the laser radar and collated with the map information. As a result, even when the recognition accuracy of the front camera is lowered, high reliability can be ensured for the position of the vehicle in the front-rear direction of the vehicle, and the relative position of the vehicle with respect to the traveling lane can be reliably detected.

It is a schematic diagram for demonstrating the vehicle in Embodiment of this invention, the photographing range of an in-vehicle camera, and the recognition area by a laser radar. It is a figure for demonstrating the vehicle control device in embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each figure, the same or corresponding parts are designated by the same reference numerals to simplify or omit the description.

FIG. 1 is a diagram for explaining a vehicle according to the present embodiment, a shooting range of an in-vehicle camera, and a recognizable area such as a structure by a laser radar. In the present embodiment, the front camera 1, the left side camera 2, and the right side camera 3 are mounted on the front, left side, and right side of the vehicle 10, respectively. Images of the front, left side, and right side of the vehicle 10 are acquired by the front camera 1, the left side camera 2, and the right side camera 3, respectively. The shooting ranges A, B, and C in FIG. 1 are ranges obtained by projecting the image ranges that can be acquired by the front camera 1, the left side camera 2, and the right side camera 3 on the plane of FIG. 1, respectively.

Although not shown, laser radars (that is, LIDAR (Light Detection and Ranging or Laser Imaging Detection and Ranging)) are mounted on the front, left side, right side, and rear of the vehicle 10. The laser radar analyzes the distance to the object and the shape of the object by measuring the scattered light with respect to the laser irradiation. The front, left, right, and rear laser radars mounted on the vehicle 10 identify the shape and relative position of road structures such as guardrails, trees, and traffic lights in front, left, right, and rear of the vehicle 10. NS. Each of the recognition areas D, E, F, and G shown in FIG. 1 is a range projected on the plane of FIG. 1 by projecting a range recognizable by each of the front, left, rear, and right laser radars. Is. The front, left, and right recognition areas D, E, and G are the shooting ranges A, B, and the shooting ranges A, B, and each of the front camera 1, the left side camera 2, and the right side camera 3, respectively. It is wider than C.

In the present embodiment, the vehicle 10 is equipped with an ECU 20 as a vehicle control device. The ECU 20 is electrically connected to various actuators mounted on the vehicle 10. The ECU 20 has a function of controlling acceleration / deceleration and steering of the vehicle 10 to execute automatic driving of the vehicle 10 by outputting a control signal having contents according to the automatic driving function to each actuator. Further, the ECU 20 also has a function of controlling acceleration / deceleration and steering of the vehicle 10 by stopping automatic driving and outputting a control signal having contents according to a driving operation input by the driver to each actuator. ing.

Further, when the image recognition accuracy of the front camera 1 is normal, the ECU 20 acquires and analyzes the image in front of the vehicle by the front camera 1 and acquires and analyzes the images from the left and right side cameras 2 and 3. do. As a result, it has a function of detecting the position of the lane division such as the white line existing on the left and right sides of the vehicle 10 and detecting the traveling lane. The ECU 20 controls the automatic driving so that the vehicle 10 travels in the traveling lane based on the detected traveling lane information during the automatic driving.

Here, since the white lines and the like in the images of the left and right side cameras 2 and 3 are large and easy to move, it is difficult to track the white lines and the like only by the side cameras 2 and 3. Further, since the images of the left and right side cameras 2 and 3 may show the seams of the road surface in parallel with the white lines and the like, the detection accuracy of the white lines and the like only by the images of the left and right side cameras 2 and 3 is not high. Therefore, in the present embodiment, the ECU 20 is configured to identify the white line or the like by using both the image of the front camera 1 and the images of the left and right side cameras 2 and 3 when detecting the traveling lane.

Further, when the image recognition accuracy of the front camera 1 is lowered due to a factor such as backlight, that is, when a malfunction of the front camera 1 is detected, the ECU 20 stops the image acquisition of the front camera 1 to the left and right. White lines and the like are detected by acquiring images from the side cameras 2 and 3 and analyzing them. Then, by acquiring and analyzing the data from the front, left, right, side, and rear laser radars, the relative position of the running vehicle 10 is confirmed, and the white lines and the like acquired by the left and right side cameras 2 and 3 are relative to each other. The control for correcting the relative position of the vehicle 10 is executed. Hereinafter, a specific description will be given.

FIG. 2 is a block diagram for explaining a function of detecting a traveling lane when a malfunction of the front camera 1 is detected among the functions of the ECU 20. As shown, the ECU 20 is connected to the front camera 1, the left side camera 2, the right side camera 3, the front laser radar 4, the left side laser radar 5, the right side laser radar 6, and the rear laser radar 7. ing.

The side image detection unit 22 acquires images of the left and right sides of the vehicle 10 by the left side camera 2 and the right side camera 3, analyzes the acquired images, and detects a white line or the like, whereby the traveling lane is detected. The relative position of the vehicle 10 in the vehicle 10 is calculated.

When a malfunction of the front camera 1 is detected, the laser radar information detection unit 24 uses the front laser radar 4, the left side laser radar 5, the right side laser radar 6, and the rear laser radar 7 to determine the road structure around the vehicle 10. Acquires the shape and position information of an object.

The collation unit 26 acquires information on the road structure around the current position of the vehicle 10 from the map database 8, and the acquired information and the shape and position information of the road structure acquired by the laser radar information detection unit 24. By collating with and, the position of the vehicle 10 on the map data is specified, and the position of the vehicle 10 in the vehicle front-rear direction is calculated. It should be noted that the acquisition of information on the road structure by the laser radar and the collation in the collation unit 26, that is, the localization, is performed more frequently than usual by actively using the road structure, etc., which is not used in the normal localization. It will be done and the frequency will be guaranteed.

The own vehicle position correction unit 28 corrects the relative position of the vehicle 10 in the traveling lane acquired by the side image detection unit 22 based on the position of the vehicle 10 calculated by the collation unit 26. That is, since the position of the vehicle 10 in the front-rear direction is correctly specified by the collation using the laser radar, the relative position of the vehicle with respect to the traveling lane calculated by the image analysis of the side cameras 2 and 3 is correctly determined based on this. to correct.

The output unit 30 calculates the acceleration / deceleration and steering force with respect to the vehicle 10 based on the position of the vehicle 10 corrected by the own vehicle position correction unit 28, and outputs a signal based on the calculation result to each actuator 32. do. As a result, the vehicle 10 is controlled so as to be reliably located in the traveling lane.

As described above, according to the present embodiment, even when the image recognition accuracy of the front camera 1 is lowered, the images of the left and right side cameras 2 and 3 are used based on the position information by the laser radars 4 to 7. The acquired relative position is corrected. As a result, the position of the vehicle 10 in the front-rear direction can be accurately specified without the image of the front camera 1. Therefore, the deviation of the images by the left and right side cameras 2 and 3 is corrected, and the vehicle in the traveling lane is correctly used. The relative positions of 10 can be specified. As a result, the automatic traveling of the vehicle 10 can be maintained even when the image of the front camera 1 cannot be used.

It should be noted that when the number, quantity, quantity, range, etc. of each element is referred to in the above-described embodiment, the reference is made except when explicitly stated or when the number is clearly specified in principle. The invention is not limited in number. 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.

1 Front camera 2 Left side camera 3 Right side camera 4 Front laser radar 5 Left side laser radar 6 Right side laser radar 7 Rear laser radar 8 Map database 10 Vehicle 22 Side image detection unit 24 Laser radar information detection unit 26 Matching unit 28 Own vehicle position correction unit 30 Output unit 32 Each actuator

Claims (1)

A front camera that acquires an image of the front of the vehicle and a side camera that acquires images of the left and right sides of the vehicle are provided, and images taken by the front camera and the side cameras are analyzed to detect and detect a traveling lane. A vehicle control device capable of calculating the relative position of the vehicle in the traveling lane.
When a decrease in recognition accuracy of the camera installed in front of the vehicle is detected,
The image acquisition of the front camera is stopped, only the image taken by the side camera is analyzed, and the relative position of the vehicle in the traveling lane is detected.
The laser radar mounted on the vehicle detects the road structure around the vehicle and the distance to the road structure, and the position of the vehicle is calculated by collating with the map information.
The relative position of the vehicle calculated by analyzing the image of the side camera is corrected according to the position of the vehicle calculated by using the laser radar.
A vehicle control device characterized in that it is configured as such.

Images