Object detecting apparatus and object detecting method

• US 8,610,620 B2
• Filed: 12/08/2009
• Issued: 12/17/2013
• Est. Priority Date: 12/08/2009
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Katoh, Masayuki
• Primary Examiner(s): Barker, Matthew M
• Application Number:
  US13/057,217
• Publication Number:
  US 20120313806A1
• Time in Patent Office: 1,470 Days
• Field of Search: 342/52
• US Class Current: 342/52
• CPC Class Codes: B60R 21/013 including means for detecting collisions, impending collisions or roll-over
• CPC Class Codes: B60R 21/0134 responsive to imminent contact with an obstacle , e.g. using radar systems
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/932 using own vehicle data, e.g. ground speed, steering wheel direction
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

An object of the present invention is to further improve the detection accuracy of a lateral position of a target in an object detecting apparatus for detecting an object by using a radar and a monocular image sensor. In the present invention, a target corresponding to a target recognized by the radar is extracted and a right edge and a left edge of the target are acquired from an image picked up by the monocular image sensor. Further, locus approximation lines, which are straight lines or predetermined curved lines for approximating loci of the right edge and the left edge, are derived for the both edges. The edge, which has a larger number of edges existing on the locus approximation line, is selected as a true edge of the target from the right edge and the left edge. The lateral position of the target is derived on the basis of the position of the selected edge.

Claims

1. An object detecting apparatus for detecting an object on the basis of target information acquired by a radar and target information acquired from an image picked up by a monocular image sensor, the image being called a monocular image detection edge, the object detecting apparatus comprising:

an edge acquiring unit configured to extract a target corresponding to a target recognized by the radar, from the image picked up by the monocular image sensor and to acquire

a set of left edges by acquiring a left edge of the extracted target every time a predetermined period of time elapses; and

a set of right edges by acquiring a right edge of the extracted target every time the predetermined period of time elapses; and

a locus approximation line deriving unit configured to derive;

for the set of left edges, a left edge locus approximation line as a straight line or predetermined curved line for approximating loci of the set of the left edges acquired by the edge acquiring unit every time the predetermined period of time elapses; and

for the set of right edges, a right edge locus approximation line as a straight line or predetermined curved line for approximating loci of the set of the right edges acquired by the edge acquiring unit every time the predetermined period of time elapses; and

a selecting unit configured to select a true edge of the target;

a lateral position deriving unit configured to derive a lateral position of the target on the basis of a position of the true edge by the selecting unit;

wherein;

the true edge of the target is selected as either

(1) a true left edge of the target in the event that a greater number of acquired left edges appear on the left edge locus approximation line than the number of acquired right edges that appear on the right edge locus approximation line;

or

(2) a true right edge of the target in the event that a greater number of acquired right edges appear on the right edge locus approximation line than the number of acquired left edges that appear on the left edge locus approximation line.

2. The object detecting apparatus according to claim 1, further comprising:

a weight applying unit configured to apply reliability weights to the right edge and the left edge acquired by the edge acquiring unit so that the edge, which is disposed nearer to a position of the target recognized by the radar, has a higher reliability; and

a reliability total value calculating unit configured to totalize a plurality of reliabilities having the weights applied by the weight applying unit for each of the right edge and the left edge to thereby calculate total values of the reliabilities for the both edges, wherein;

the selecting unit selects, as the true edge of the target, the edge which has the larger total value of the reliabilities calculated by the reliability total value calculating unit, from the right edge and the left edge, if the number of the edges existing on the locus approximation line is identical between the right edge and the left edge acquired by the edge acquiring unit.

3. The object detecting apparatus according to claim 2, wherein the lateral position deriving unit includes:

a locus predicting unit configured to predict a future locus of the edge selected as the true edge by the selecting unit; and

a collision position predicting unit configured to predict a collision position between the target and a vehicle as a position at which a distance between the edge and the vehicle in the front-back direction is zero on the basis of the future locus of the edge predicted by the locus predicting unit, wherein;

a lateral position of a center in a lateral direction of the target, which is to be provided at the collision position, is derived on the basis of a position at the collision position of the edge selected as the true edge by the selecting unit.

4. The object detecting apparatus according to claim 2, wherein:

the lateral position deriving unit includes a lateral width estimating unit configured to estimate a lateral width of the target; and

a position, which is deviated toward the other edge by ½

of the lateral width of the target estimated by the lateral width estimating unit from the position of the edge selected as the true edge by the selecting unit, is derived as a lateral position of a center in a lateral direction of the target.

5. The object detecting apparatus according to claim 1, wherein the lateral position deriving unit includes:

a locus predicting unit configured to predict a future locus of the edge selected as the true edge by the selecting unit; and

a collision position predicting unit configured to predict a collision position between the target and a vehicle as a position at which a distance between the edge and the vehicle in the front-back direction is zero on the basis of the future locus of the edge predicted by the locus predicting unit, wherein;

a lateral position of a center in a lateral direction of the target, which is to be provided at the collision position, is derived on the basis of a position at the collision position of the edge selected as the true edge by the selecting unit.

6. The object detecting apparatus according to claim 1, wherein:

the lateral position deriving unit includes a lateral width estimating unit configured to estimate a lateral width of the target; and

a position, which is deviated toward the other edge by ½

of the lateral width of the target estimated by the lateral width estimating unit from the position of the edge selected as the true edge by the selecting unit, is derived as a lateral position of a center in a lateral direction of the target.

7. An object detecting method for detecting an object on the basis of target information acquired by a radar and target information acquired from an image picked up by a monocular image sensor, the image being called a monocular image detection edge, the object detecting method executed by an electronic control unit and comprising:

an edge acquiring step of extracting a target corresponding to a target recognized by the radar, from the image picked up by the monocular image sensor and acquiring;

a set of left edges by acquiring a left edge of the extracted target every time a predetermined period of time elapses; and

a set of right edges by acquiring a right edge of the extracted target every time the predetermined period of time elapses; and

a locus approximation line deriving step of deriving;

for the set of left edges, a left edge locus approximation line as a straight line or predetermined curved line for approximating loci of the set of the left edges acquired in the edge acquiring step every time a predetermined period of time elapses; and

for the set of right edges, a right edge locus approximation line as a straight line or predetermined curved line for approximating loci of the set of the right edges acquired by the edge acquiring unit every time the predetermined period of time elapses; and

a selecting step of selecting a true edge of the target; and

a lateral position deriving step of deriving a lateral position of the target on the basis of a position of the true edge selected in the selecting step;

wherein;

the true edge of the target is selected as either;

(1) a true left edge of the target in the event that a greater number of acquired left edges appear on the left edge locus approximation line than the number of acquired right edges that appear on the right edge locus approximation line;

or

(2) a true right edge of the target in the event that a greater number of acquired right edges appear on the right edge locus approximation line than the number of acquired left edges that appear on the left edge locus approximation line.

8. The object detecting method according to claim 7, further comprising:

a weight applying step of applying reliability weights to the right edge and the left edge acquired in the edge acquiring step so that the edge, which is disposed nearer to a position of the target recognized by the radar, has a higher reliability; and

a reliability total value calculating step of totalizing a plurality of reliabilities having the weights applied in the weight applying step for each of the right edge and the left edge to thereby calculate total values of the reliabilities for the both edges, wherein;

the edge, which has the larger total value of the reliabilities calculated in the reliability total value calculating step, is selected as the true edge of the target from the right edge and the left edge in the selecting step, if the number of the edges existing on the locus approximation line is identical between the right edge and the left edge acquired in the edge acquiring step.