Obstacle Detection Apparatus

• US 20080136612A1
• Filed: 11/16/2007
• Published: 06/12/2008
• Est. Priority Date: 11/16/2006
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Otsuka, Yuji, Muramatsu, Shoji, Machii, Kimiyoshi, Shima, Takeshi, Higuchi, Mirai
• Granted Patent: US 7,782,179 B2
• Time in Patent Office: Days
• US Class Current: 340/435
• CPC Class Codes: G01S 11/12 using electromagnetic waves other than radio waves
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Abstract

Disclosed is an obstacle detection apparatus to detect more accurately a moving obstacle present around a motor vehicle. With this system, desired feature quantities of an obstacle present in a coverage of a first sensor are calculated and then stored into a storage device, and when the obstacle moves from the coverage of the first sensor to a coverage of a second sensor, a relative distance to the obstacle is calculated from the above-stored feature quantities of the obstacle and from the desired feature quantities of the obstacle, obtained by analyzing the recognition results of the second sensor that have been processed by a processor which processes the recognition results.

Claims

1. An obstacle detection apparatus comprising:

a first processor which processes results of recognition with a first sensor;

a second processor which processes results of recognition with a second sensor;

an arithmetic unit constructed such that by analyzing the recognition results which has been processed by said first processor, said arithmetic unit calculates desired feature quantities of an obstacle present in a coverage of the first sensor and stores the calculated feature quantities into a storage device, and such that when the obstacle moves from the coverage of the first sensor to a coverage of the second sensor, said arithmetic unit calculates a relative distance of the obstacle, from the feature quantities of the obstacle which have been stored into the storage device, and from the desired feature quantities of the obstacle which have been obtained by analyzing the recognition results of the second sensor, processed by said second processor; and

an interface unit for outputting the relative distance data acquired by said arithmetic unit.

2. The obstacle detection apparatus according to claim 1, wherein:

said first processor processes an image detected by a stereo camera which images a rear region of a vehicle;

said second processor processes an image detected by a monocular camera which images a rear-quarter region of the vehicle; and

said arithmetic unit is constructed such that by analyzing the image which has been processed by said first processor, said arithmetic unit calculates vehicle width of a following vehicle present in a coverage of the stereo camera and stores the calculated vehicle width into a storage device, and such that when the following vehicle moves from the coverage of the stereo camera to a coverage of the monocular sensor, said arithmetic unit calculates a distance between the following vehicle and host vehicle, from the vehicle width of the following vehicle which has been stored into the storage device, and from the vehicle width of the following vehicle which has been obtained by analyzing the image processed by said second processor.

3. The obstacle detection apparatus according to claim 2, wherein:

if said arithmetic unit judges that the vehicle detected by the monocular camera is a vehicle which has not passed through the coverage of the stereo camera, said arithmetic unit calculates the distance between the following vehicle and the host vehicle, from feature quantities of the following vehicle which have been obtained from the image processed by said second processor.

4. The obstacle detection apparatus according to claim 2, wherein:

said arithmetic unit calculates the distance between the following vehicle and the host vehicle, from feature quantities of the following vehicle which have been obtained from the image processed by said first or second processor, and if a differential distance is less than a threshold value, said arithmetic unit judges that the following vehicle is the same as the vehicle having the feature quantities stored within the storage device, and updates the feature quantities stored within the storage device.

5. The obstacle detection apparatus according to claim 2, wherein:

the storage device retains at least either the vehicle width of the vehicle detected by said first processor, the vehicle width of the vehicle detected by said second processor, a relative position to the host vehicle, detection time, a relative speed, or the vehicle width in the image.

6. The obstacle detection apparatus according to claim 2, wherein:

said arithmetic unit includes a host vehicle lane following vehicle detector which detects the following vehicle present in the same lane as that of the host vehicle, by analyzing the image processed by said first processor, a first passing-lane following vehicle detector which detects a following vehicle present in a passing lane for the same lane as that of the host vehicle, by analyzing the image processed by said first processor, and a second passing-lane following vehicle passage detector which defects the following vehicle present in the passing lane, by analyzing the image processed by said second processor.

7. The obstacle detection apparatus according to claim 6, wherein:

said arithmetic unit includes a lane markings recognizer to recognize a lane marking on a road by analyzing the image processed by said first processor, and recognizes the traveling lane of the following vehicle by a position of gravitational center thereof on a screen and a relative position of the lane marking recognized by the lane markings recognizer.

8. The obstacle detection apparatus according to claim 1, wherein:

said first processor processes an image detected by a first monocular camera which images a rear region of a vehicle;

said second processor processes an image detected by a second monocular camera which images a rear-quarter region of the vehicle; and

said arithmetic unit is constructed such that by analyzing the image which has been processed by said first processor, said arithmetic unit calculates vehicle width or height of a following vehicle present in a coverage of the first monocular camera and stores the calculated vehicle width or height into a storage device, and such that when the following vehicle moves from the coverage of the first monocular camera to a coverage of the second monocular sensor, said arithmetic unit calculates a distance between the following vehicle and host vehicle, from the vehicle width or height of the following vehicle which has been stored into the storage device, and from the vehicle width or height of the following vehicle which has been obtained by analyzing the image processed by said second processor.

9. The obstacle detection apparatus according to claim 8, wherein:

said arithmetic unit detects a current traveling lane of the host vehicle from the image acquired by the first monocular camera, estimates vehicle width of the vehicle by using not only the vehicle width estimated from both a distance between the lane markings and the kind of road on which the vehicle is traveling, but also width of the vehicle in the image thereof; and

said second image processor calculates a distance to the vehicle in the image of the second monocular camera by using the vehicle width.

10. The obstacle detection apparatus according to claim 1, wherein:

said first processor processes results of detection with a laser radar which detects a moving obstacle present at vehicle rear;

said second processor processes an image detected by a monocular camera which images a rear-quarter region of the vehicle; and

said arithmetic unit is constructed such that by analyzing the results which has been processed by said first processor, said arithmetic unit calculates feature quantities of a following vehicle present in a coverage of the laser radar and stores the calculated feature quantities into a storage device, and such that when the following vehicle moves from the coverage of the laser radar to a coverage of the monocular camera, said arithmetic unit calculates a distance between the following vehicle and host vehicle, from the feature quantities of the following vehicle which have been stored into the storage device, and from the feature quantities of the following vehicle which have been obtained by analyzing the image processed by said second processor.

11. The obstacle detection apparatus according to claim 10, wherein:

said arithmetic unit calculates an absolute position of the detected vehicle from a relative position thereof and a position of the host vehicle, then calculates a distance between the detected vehicle and a link constituting a road on which the host vehicle is traveling, and if the distance is less than a threshold value, judges the detected vehicle to be traveling in the same lane as that of the host vehicle, or if the distance is equal to or greater than the threshold value, discriminates the traveling lane of the detected vehicle by a positional relationship between the vehicle and the link.