Position estimation for ground vehicle navigation based on landmark identification/yaw rate and perception of landmarks

• US 8,301,374 B2
• Filed: 08/25/2009
• Issued: 10/30/2012
• Est. Priority Date: 08/25/2009
• Status: Active Grant

• Current Assignee: Southwest Research Institute
• Original Assignee: Southwest Research Institute
• Inventors: Surampudi, Bapiraju, Steiber, Joe
• Primary Examiner(s): Tarcza, Thomas
• Assistant Examiner(s): Khatib, Rami
• Application Number:
  US12/547,213
• Publication Number:
  US 20110054791A1
• Time in Patent Office: 1,162 Days
• Field of Search: 701 23- 28, 701400-541, 701300-302, 340/901, 340/905, 340933-943, 3409951-996, 340/686.6
• US Class Current: 701/438
• CPC Class Codes: G01C 21/005 with correlation of navigation data from several sources, e.g. map or contour matching G01C21/30 takes precedence
• CPC Class Codes: G01C 21/12 executed aboard the object being navigated; Dead reckoning
• CPC Class Codes: G01C 21/30 Map- or contour-matching

Abstract

A landmark-based method of estimating the position of an AGV (autonomous ground vehicle) or conventional vehicle, which has an onboard database of landmarks and their locations coordinates. During travel, the AGV looks for and identifies landmarks. It navigates according to position estimations that are based on measured yaw rate and speed. When a landmark s encountered and recognized, its true location coordinates are looked up from the database. These true coordinates are then used to enhance position estimation accuracy between landmarks.

Claims

1. A method of estimating a position of an autonomous ground vehicle during travel along a route, the ground vehicle having locomotion equipment and steering equipment, comprising:

storing in a database, a set of landmarks and landmark location coordinates for each landmark;

initializing position with known location coordinates;

receiving current yaw rate and vehicle speed data;

using visual perception equipment and the database to detect and identify landmarks;

when no landmark is identified, using a processing unit to estimate current position data of the ground vehicle, the dead reckoning process performed by calculating latitude and longitude coordinates and a heading based on the current yaw rate and vehicle speed data;

when a landmark is identified, using

its landmark location coordinates to correct the current position data;

wherein at least some of the landmarks are curbs, each curb at an intersection of two roadways, each of the roadways having two parallel boundary lines, and each curb having landmark location coordinates defined by the intersection of a boundary line of each of the roadways;

wherein an intersection is identified by identifying at least two curbs, and at least one of the two curbs is used to provide landmark location coordinates to correct the current position data; and

using the current position data to provide input to the steering equipment.

2. The method of claim 1, wherein the visual perception equipment is LIDAR, Optical Camera, or Sonar.

3. The method of claim 1, wherein the landmark location coordinates are GPS coordinates.

4. The method of claim 1, wherein the vehicle uses live GPS-based navigation when available and the method is used when live GPS navigation is not available.

5. The method of claim 1, wherein the vehicle is always navigated without a live GPS signal.

6. The method of claim 1, wherein the method is performed by said processing unit onboard the vehicle.

7. The method of claim 1, wherein at least one landmark is a lane marking.

8. The method of claim 1, wherein at least one landmark is a lane width.

9. The method of claim 1, wherein the step of using landmark location coordinates to correct the current position data is performed in part by providing offsets to the landmark location coordinates to adjust for the difference between the vehicle'"'"'s actual position and the landmark location coordinates, and wherein the offsets are calculated by comparing an image of an intersection from the visual perception equipment to a stored image of the intersection.

10. A control unit for estimating a position of an autonomous ground vehicle during travel along a route, the vehicle having visual perception equipment, a yaw rate sensor, and vehicle speed sensor, comprising:

a landmark database for storing a set of landmarks and each landmark'"'"'s location coordinates;

a processing unit programmed to perform the following tasks;

to initialize position with known location coordinates;

to receive current yaw rate and vehicle speed data;

to use the visual perception equipment and the database to look for and to identify landmarks;

when no landmark is identified, estimating the current position data of the vehicle by calculating location coordinates and a heading based on the current yaw rate and the vehicle speed data;

when a landmark is identified, to use its landmark coordinates to current the current position data;

to use the landmark location coordinates to correct the vehicle'"'"'s current position data;

wherein at least some of the landmarks are curbs, each curb at an intersection of two roadways, each of the roadways having two parallel boundary lines, and each curb having landmark location coordinates defined by the intersection of a boundary line of each of the roadways;

wherein an intersection is identified by identifying at least two curbs, and at least one of the two curbs is used to provide landmark location coordinates to correct the current position data.

11. The control unit of claim 10, wherein the visual perception equipment is LIDAR or optical camera or sonar.

12. The control unit of claim 10, wherein the landmark location coordinates are GPS coordinates.

13. The control unit of claim 10, wherein the landmark database and processing unit are onboard the vehicle.

14. The control unit of claim 10, Wherein the processing unit is in data communications with a remote server programmed to more landmark data and to extract features of upcoming portions of the route.

15. The control unit of claim 10, wherein at least one landmark is a lane marking.

16. The control unit of claim 10, wherein at least one landmark is a lane width.

17. The system of claim 10, wherein the task of using landmark location coordinates to correct the current position data is performed in part by providing offsets to the landmark location coordinates to adjust for the difference between the vehicle'"'"'s actual position and the landmark location coordinates, and wherein the offsets are calculated by comparing an image of an intersection from the visual perception equipment to a stored image of the intersection.