Road environment recognition device and method of recognizing road environment

• US 8,694,236 B2
• Filed: 04/24/2007
• Issued: 04/08/2014
• Est. Priority Date: 05/17/2006
• Status: Active Grant

• Current Assignee: DENSO Corporation
• Original Assignee: DENSO Corporation
• Inventors: Takagi, Kiyokazu
• Primary Examiner(s): Black, Thomas
• Assistant Examiner(s): Kong, Sze-Hon
• Application Number:
  US11/789,178
• Publication Number:
  US 20080040039A1
• Time in Patent Office: 2,541 Days
• Field of Search: 342 70- 72, 701300-302, 701/32.3, 701/408, 701/514, 701/519, 701/520, 340/994.24
• US Class Current: 701/300
• CPC Class Codes: G01S 17/931 of land vehicles

Abstract

A radar unit emits beams, and receives a reflection beam reflected by an object. A position of the object relative to a vehicle and an attribute of the object are recognized based on the emitted beams and the reflection beam. A coordinate position of the vehicle in an absolute coordinate is calculated based on a traveling amount of the vehicle, and a coordinate position of the object is calculated based on the calculated position of the vehicle and the position of the object relative to the vehicle. A road environment of the vehicle is recognized based on the coordinate positions and the attribute of the object.

Claims

1. A road environment recognition device comprising:

a radar unit on a subject vehicle, wherein the radar unit emits a plurality of beams in a predetermined angle range in a width direction and a height direction of the subject vehicle, and receives a plurality of reflection beams reflected by objects;

an object recognizing element recognizing a position of one object relative to the subject vehicle and an attribute of the one object based on a first portion of the emitted plurality of beams and the reflection beam;

a lane marker recognizing element recognizing a position of a lane mark relative to the subject vehicle based on a second portion of the emitted plurality of beams different than the first portion of the emitted plurality of beams and the reflection beam;

a traveling amount measuring element measuring a traveling amount of the subject vehicle;

a coordinate calculating element repeatedly calculating an absolute coordinate of the subject vehicle in an absolute coordinate system having a predetermined point as an origin point based on the traveling amount measured by the traveling amount measuring element, repeatedly calculating an absolute coordinate of the one object based on the calculated absolute coordinate of the subject vehicle and the position of the one object relative to the subject vehicle and repeatedly calculating an absolute coordinate of the position of the lane mark based on the calculated absolute coordinate of the subject vehicle and the position of the lane mark relative to the subject vehicle; and

a road environment recognizing element memorizing a plurality of data points of the calculated absolute coordinates of the subject vehicle, memorizing a plurality of data points of the calculated absolute coordinates of the one object and memorizing a plurality of data points of the calculated absolute coordinates of the position of the lane mark, and recognizing a road environment of the subject vehicle by integrating the memorized plurality of data points of the absolute coordinates and the attribute of the one object with the memorized plurality of data points of the absolute coordinates of the position of the lane mark;

wherein

the lane mark representing a boundary of a lane on which the subject vehicle is driven;

the radar unit emits the first portion of the plurality of beams toward a forward area of the subject vehicle, and the second portion of the plurality of beams is radiated toward a road face;

the radar unit receives light reflected by the lane mark, when the second portion of the plurality of beams radiated toward the road face is reflected by the lane mark; and

the road environment recognizing element recognizes a lane change of the subject vehicle based on a change of the absolute coordinate of the subject vehicle relative to the absolute coordinate of the lane mark in the absolute coordinate system;

wherein present data is unified with past data by updating all of the plurality of data points of the absolute coordinates of the one object detected in a past detection point in accordance with a change in position of the subject vehicle.

2. The road environment recognition device according to claim 1, wherein

the object recognizing element recognizes the attribute of the one object based on at least shape information provided by a portion of the plurality of the reflection beams, and an intensity of the portion of the plurality of the reflection beams.

3. The road environment recognition device according to claim 1, wherein the road environment recognition device further recognizes a roadside object positioned at one side of a road,

the road environment recognizing element recognizes a number of lanes on the road based on a lane width provided from a position of the roadside object and a position of the lane mark in the absolute coordinate system.

4. The road environment recognition device according to claim 3, wherein

the road environment recognizing element recognizes the lane on which the subject vehicle is driven based on a position of the subject vehicle in the absolute coordinate system.

5. The road environment recognition device according to claim 3, further comprising:

a road curvature detecting element detecting a curvature of the road on which the subject vehicle is driven, wherein

the road environment recognizing element distinguishes the recognized roadside object between a left side roadside object and a right side roadside object in accordance with the curvature of the road.

6. The road environment recognition device according to claim 1, wherein the road environment recognizing element further recognizes a forward vehicle driven in a forward area of the subject vehicle,

the road environment recognizing element recognizes a number of lanes on the road based on a relationship between a driving trajectory of the forward vehicle and a driving trajectory of the subject vehicle in the absolute coordinate system.

7. The road environment recognition device according to claim 1, wherein the road environment recognizing element further recognizes a roadside object in a forward area of the subject vehicle,

the road environment recognizing element recognizing the road environment based on the memorized absolute coordinates of the roadside object and another memorized absolute coordinate of another roadside object, which was detected in a past detection period.

8. The road environment recognition device according to claim 1, wherein:

the road environment recognizing element recognizes that the lane change has started when a distance between a center position of the vehicle and a center of the lane becomes larger than a predetermined threshold value; and

the road environment recognizing element recognizes that the lane change is finished when the entire vehicle crosses the lane mark.

9. The road environment recognition device according to claim 1, wherein the road environment recognizing element forms a trajectory of the one object based on all of the memorized plurality of data points of the absolute coordinates of the one object.

10. The road environment recognition device according to claim 1, wherein continuation from the past data to the present data is secured only by converting latest data into the absolute coordinate.

11. The road environment recognition device according to claim 1, wherein the memorized plurality of data points of the calculated absolute coordinates of the one object are connected into segments to provide a trajectory of the one object.

12. A method of recognizing a road environment comprising:

emitting a plurality of beams from a radar unit mounted on a vehicle in a predetermined angle range in a width direction and a height direction of the vehicle and the plurality of beams are reflected by objects as a plurality of reflection beams;

recognizing a position of one object relative to the vehicle and an attribute of the one object based on a first portion of the emitted plurality of beams and the reflection beam reflected by the one object;

recognizing a position of a lane mark relative to the vehicle based on a second plurality of the emitted plurality of beams different from the first portion and the reflection beam;

measuring a traveling amount of the vehicle;

repeatedly calculating an absolute coordinate of the vehicle in an absolute coordinate system having a predetermined point as an origin point based on the measured traveling amount, repeatedly calculating an absolute coordinate of the one object based on the calculated absolute coordinate of the vehicle and the position of the object relative to the vehicle and repeatedly calculating an absolute coordinate of the position of the lane mark based on the calculated absolute coordinate of the vehicle and the position of the lane mark relative to the vehicle; and

memorizing a plurality of data points of the calculated absolute coordinate of the vehicle, a plurality of data points of the one object and a plurality of data points of the position of the lane mark;

recognizing a road environment of the vehicle based on the memorized plurality of data points of the absolute coordinates of the vehicle and by integrating the memorized plurality of data points of the absolute coordinate of the one object, and the attribute of the one object with the memorized plurality of data points of the absolute coordinates of the position of the lane mark;

wherein,

the lane mark representing a boundary of a lane on which the vehicle is driven;

the radar unit emits the first portion of the plurality of beams toward a forward area of the vehicle, and the second portion of the plurality of beams is radiated toward a road face;

the radar unit receives light reflected by the lane mark, when the second portion of the plurality of beams radiated toward the road face is reflected by the lane mark;

the road environment recognizing element recognizes a lane change of the vehicle based on a change of the absolute coordinate of the vehicle relative to the absolute coordinate of the lane mark in the absolute coordinate system;

wherein present data is unified with past data by updating all of the plurality of data points of the absolute coordinates of the one object detected in a past detection point in accordance with a change in position of the vehicle.

13. The method of recognizing a road environment according to claim 12, further comprising:

recognizing a roadside object in a forward area of the vehicle, and

memorizing absolute coordinates of the roadside object;

wherein

the step of recognizing the road environment of the vehicle is performed based on the memorized absolute coordinates of the roadside object and another memorized absolute coordinate of a past roadside object, which was detected in a past detection period.

14. The method of recognizing a road environment according to claim 12, wherein:

the road environment recognizing element recognizes that the lane change has started when a distance between a center position of the vehicle and a center of the lane becomes larger than a predetermined threshold value; and

the road environment recognizing element recognizes that the lane change is finished when the entire vehicle crosses the lane mark.

15. The method of recognizing a road environment according to claim 12, further comprising forming a trajectory of the one object based on all of the memorized plurality of data points of the absolute coordinates of the one object.

16. The road environment recognition device according to claim 12, further comprising connecting the memorized plurality of data points of the calculated absolute coordinates of the one object into segments to provide a trajectory of the one object.