Target recognition and localization methods using a laser sensor for wheeled mobile robots

Patent metadata

Field	Value
Publication number	US-9739616-B2
Application number	US-201214123062-A
Country	US
Kind code	B2
Filing date	Jun 1, 2012
Priority date	Jun 2, 2011
Publication date	Aug 22, 2017
Grant date	Aug 22, 2017

Abstract

A localization scheme and method using a laser sensor for indoor wheeled mobile robots (IWMR), which need to localize themselves for working autonomously, is provided. In this method, a laser sensor moves inside an onboard guide way and its distance measurements are used to robustly detect and recognize a unique target based on edge detection and pattern recognition techniques. From the distance measurements with respect to the recognized target, a kinematic model is developed to determine the IWMR orientation and location in the global co-ordinates (in 2-D). Such target recognition and localization methods are validated with experimental results.

First claim

The invention claimed is:

1. A computer implemented method for localizing a wheeled robot/vehicle using a laser sensor, comprising:
scanning an angular section of a surrounding environment of the wheeled robot/vehicle by the laser sensor, wherein the laser sensor is mounted on a rotary actuator of the wheeled robot/vehicle;
identifying a target, wherein identifying the target includes detecting edges on the target and recognizing a pattern of the target based on the scanning by the laser sensor, wherein detecting the edges on the target is based on laser sensor measurements and rotary actuator angular positions of the rotary actuator, and wherein detecting the edges on the target includes identifying two edges that have a minimum length equal to a minimum depth of the target;
transforming the laser sensor measurements and the rotary actuator angular positions into vehicular coordinates, wherein the vehicular coordinates are associated to the target as identified from the laser sensor and derived from time instant points corresponding to the two edges, wherein the time instant points are joined in a straight line;
determining global coordinates for the vehicular coordinates based on a slope of the straight line; and
controlling the wheeled robot/vehicle to travel autonomously based on the global coordinates.

2. The computer implemented method of claim 1 wherein the determined global coordinates are compared with templates of the target to confirm that the target has been recognized.

3. The computer implemented method of claim 2 wherein comparing determined global coordinates with templates includes translating the templates that are defined in global coordinates into laser coordinates and minimizing Hausdorff distances between the determined global coordinates and the templates.

4. The computer implemented method of claim 1 further including:
generating templates for the target, wherein templates for the target include three different templates that are each utilized based on the location of the wheeled robot/vehicle.

5. The computer implemented method of claim 1 wherein controlling the wheeled robot/vehicle includes:
determining an orientation and location of the wheeled robot/vehicle based on the global coordinates.

6. The computer implemented method of claim 4 wherein each different template of the three different templates represent three different zones in front of the target, wherein the wheeled robot/vehicle is located within one of the three different zones.

7. The computer implemented method of claim 1 further including reducing the influence of noise in the laser sensor measurements on the determined global coordinates.

8. The computer implemented method of claim 7 further including approximating the noise of the laser sensor measurements by a white Gaussian noise.

9. The computer implemented method of claim 1 wherein the laser sensor has rotary and linear motion capabilities.

Assignees

• Wang Junmin
• Govindarajan Madhu Soodhanan
• Post Ii James W
• Fox Andrew John
• Honda Motor Co Ltd

Inventors

• Wang Junmin
• Govindarajan Madhu Soodhanan
• Post Ii James W
• Fox Andrew John

Classifications

• G06T12/10

  Image preprocessing, e.g. calibration, positioning of sources or scatter correction · CPC title

• G05D1/024

  in combination with a laser (lasers per se H01S) · CPC title

• G01C21/206 Primary

  specially adapted for indoor navigation · CPC title

• G06T11/005

  Physics · mapped topic

• G01B7/004

  for measuring coordinates of points · CPC title