Deep lane navigation system for automatic guided vehicles

• US 9,046,893 B2
• Filed: 05/31/2012
• Issued: 06/02/2015
• Est. Priority Date: 05/31/2011
• Status: Active Grant

• Current Assignee: John Bean Technologies Corporation
• Original Assignee: John Bean Technologies Corporation
• Inventors: Douglas, Barry, Repko, Michael
• Primary Examiner(s): Nguyen, Bao Long T
• Application Number:
  US13/485,492
• Publication Number:
  US 20120330492A1
• Time in Patent Office: 1,097 Days
• US Class Current: 1/1
• CPC Class Codes: G05D 1/0236 in combination with a laser lasers per se H01S
• CPC Class Codes: G05D 1/024 in combination with a laser lasers per se H01S
• CPC Class Codes: Y10S 901/01 Mobile robot

Abstract

A method for automatically guiding a vehicle along at least a first of a plurality of rows comprising predefined centerlines. At least a second of the rows comprises a number of objects of known dimensions positioned at known locations along the centerline thereof. The method comprises the steps of scanning the objects from the vehicle, generating a set of data points representative of the locations of the objects relative to the vehicle, determining a sensed position and heading of the vehicle from the data points, comparing the sensed position and heading of the vehicle to a position and direction of the first row centerline, and generating offset and heading errors for the vehicle based on differences between the sensed position and heading of the vehicle and the position and direction of the first row centerline.

Claims

1. A method for automatically guiding a vehicle relative to a centerline extending along at least a first of a plurality of navigatable rows, each row comprising predefined centerlines, at least a second of said rows comprising a number of movable objects positioned at locations along the centerline of at least a second of said rows, the method comprising:

scanning from the vehicle the movable objects positioned at locations along the at least second of said rows;

generating a set of data points representative of the locations of the movable objects relative to the vehicle;

analyzing the data points to determine the true location of the first row being traveled by the vehicle;

determining a sensed position and heading of the vehicle relative to the centerline of the first row;

comparing the sensed position and heading of the vehicle to a position and direction of the first row centerline;

generating offset and heading errors for the vehicle based on differences between the sensed position and heading of the vehicle and the position and direction of the first row centerline; and

based on the generated offset and heading errors, altering the direction of travel of the vehicle.

2. The method of claim 1, wherein the vehicle comprises a main navigation laser and at least one bumper laser, and wherein the scanning step is performed by the bumper laser.

3. The method of claim 2, wherein the vehicle comprises a front bumper laser and a rear bumper laser, and wherein the scanning step is performed by both the front and rear bumper lasers.

4. The method of claim 1, wherein the step of determining the sensed position and heading of the vehicle comprises determining a sensed lane edge of the movable objects from the data points.

5. The method of claim 4, wherein the step of determining the sensed lane edge comprises:

clustering the data points corresponding to each movable object; and

generating a line segment for each cluster of data points;

wherein each line segment is representative of a sensed edge of the movable objects.

6. The method of claim 5, wherein the step of determining the sensed lane edge further comprises merging the line segments into the sensed lane edge.

7. The method of claim 4, wherein the step of comparing the sensed position and heading of the vehicle to a position and direction of the first row centerline comprises:

determining a virtual lane edge representative of the actual edge of the movable objects; and

comparing the sensed lane edge with the virtual lane edge.

8. The method of claim 7, wherein the virtual lane edge is determined based upon the known locations and dimensions of the movable objects.