Method and system for recognizing target parking location of vehicle by detecting range data

• US 8,260,001 B2
• Filed: 10/24/2008
• Issued: 09/04/2012
• Est. Priority Date: 10/26/2007
• Status: Active Grant

• Current Assignee: HL Klemove Corporation
• Original Assignee: Mando Corporation
• Inventors: Jung, Ho-gi
• Primary Examiner(s): Werner, Brian P
• Assistant Examiner(s): ANSARI, TAHMINA N
• Application Number:
  US12/288,928
• Publication Number:
  US 20090146842A1
• Time in Patent Office: 1,411 Days
• Field of Search: None
• US Class Current: 382/104
• CPC Class Codes: B60W 2420/408 Radar; Laser, e.g. lidar
• CPC Class Codes: B60W 30/095 Predicting travel path or likelihood of collision
• CPC Class Codes: B60W 50/14 Means for informing the driver, warning the driver or prompting a driver intervention
• CPC Class Codes: G01S 17/42 Simultaneous measurement of distance and other co-ordinates indirect measurement G01S17/46
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: G01S 2013/9314 Parking operations
• CPC Class Codes: G01S 7/4802 using analysis of echo signal for target characterisation; Target signature; Target cross-section
• CPC Class Codes: G06F 18/2323 based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
• CPC Class Codes: G06V 10/7635 based on graphs, e.g. graph cuts or spectral clustering
• CPC Class Codes: G06V 20/586 of parking space

Abstract

Disclosed is a method and a system for recognizing a target parking location of a vehicle. The system includes a scanning laser radar for detecting range data reflected and directed from a rear surface of the vehicle; a preprocessing unit for preprocessing the range data to extract effective clusters; a corner recognition unit for extracting obstacle corners from the effective clusters; and a parking location setting unit for recognizing a main reference corner, which is closest to a subjective vehicle and has an adjacent space satisfying an available parking space condition, and a sub reference corner, which exists toward the adjacent space in a direction opposite to a vehicle width direction from the main reference corner, and corresponds to a point on a closest effective cluster within a distance between a vehicle width and a vehicle length, from the obstacle corners to set a target parking location.

Claims

1. A system for recognizing a target parking location of a vehicle, the system comprising:

a scanning laser radar for detecting range data reflected and directed from a rear surface of the vehicle;

a preprocessing unit for preprocessing the range data to extract effective clusters of the range data by removing clusters of which lengths are less than a threshold, clusters being continuous range data;

a corner recognition unit for extracting obstacle corners from the effective clusters; and

a parking location setting unit for recognizing a main reference corner and a sub reference corner from the obstacle corners to set a target parking location, the main reference corner being an obstacle corner closest to a subject vehicle and having an adjacent space satisfying an available parking space condition, the sub reference corner being an obstacle corner existing toward the adjacent space in a direction opposite to a vehicle width direction from the main reference corner and being a point on a closest effective cluster within a distance between a vehicle width and a vehicle length.

2. The system as claimed in claim 1, wherein the obstacle corner is a corner of an L-shaped effective cluster or a corner of an effective cluster having a combined shape of an elliptical arc and a straight line.

3. The system as claimed in claim 1, wherein the available parking space condition is satisfied when an obstacle object exists within a distance between the vehicle width and the vehicle length within a predetermined region of interest in the direction opposite to the vehicle width direction from the obstacle corner, and when no obstacle object exists within the distance of the vehicle length within the predetermined region of interest in a direction opposite to a vehicle length direction from the obstacle corner.

4. The system as claimed in claim 1, wherein the sub reference corner is one of an occlusion and the obstacle corner.

5. The system as claimed in claim 1, wherein the target parking location is set by projecting the sub reference corner onto a point on a straight line extending from the main reference corner in the vehicle length direction, designating a line segment extending from the point of projection to a point of the main reference corner, the point lying toward the outside of the adjacent space, in parallel with the vehicle width direction as an outward border, and locating a line segment extending through the center of the outward border in parallel with the vehicle length direction at the center of the target parking location.

6. A method for recognizing a target parking location of a vehicle by a system having a scanning laser radar, a preprocessing unit, a corner recognition unit, and a parking location setting unit, the method comprising:

detecting range data by the scanning laser radar;

preprocessing the range data, by the preprocessing unit, to extract effective clusters of the range data by removing clusters of which lengths are less than a threshold, clusters being continuous range data;

extracting obstacle corners from the clusters by the corner recognition unit; and

recognizing a main reference corner and a sub reference corner from the obstacle corners by the parking location setting unit to set a target parking location, the main reference corner being an obstacle corner closest to a subject vehicle and having an adjacent space satisfying an available parking space condition, the sub reference corner being an obstacle corner existing toward the adjacent space in a direction opposite to a vehicle width direction from the main reference corner and being a point on a closest effective cluster within a distance between a vehicle width and a vehicle length.

7. The method as claimed in claim 6, wherein the preprocessing comprises:

removing ineffective data and isolated data from the range data to extract effective range data;

recognizing occlusions from the effective range data; and

removing small clusters from the effective range data.

8. The method claimed in claim 6, wherein the obstacle corner is a corner of an L-shaped effective cluster or a corner of an effective cluster having a combined shape of an elliptical arc and a straight line.

9. The method as claimed in claim 6, wherein the available parking space condition is satisfied when an obstacle object exists within a distance between the vehicle width and the vehicle length within a predetermined region of interest in the direction opposite to the vehicle width direction from the obstacle corner, and when no obstacle object exists within the distance of the vehicle length within the predetermined region of interest in a direction opposite to a vehicle length direction from the obstacle corner.

10. The method as claimed in claim 6, wherein the sub reference corner is one of an occlusion and the obstacle corner.

11. The method as claimed in claim 6, wherein the target parking location is set by projecting the sub reference corner onto a point on a straight line extending from the main reference corner in the vehicle length direction, designating a line segment extending from the point of projection to a point of the main reference corner, the point lying toward the outside of the adjacent space, in parallel with the vehicle width direction as an outward border, and locating a line segment extending through the center of the outward border in parallel with 15 the vehicle length direction at the center of the target parking location.