MOBILE ROBOT AND SIMULTANEOUS LOCALIZATION AND MAP BUILDING METHOD THEREOF

• US 20120106828A1
• Filed: 08/26/2011
• Published: 05/03/2012
• Est. Priority Date: 11/03/2010
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Yoon, Suk June, Roh, Kyung Shik, Hyung, Seung Yong, Ahn, Sung Hwan, Hwang, Hyo Seok
• Granted Patent: US 9,058,521 B2
• Time in Patent Office: Days
• US Class Current: 382/153
• CPC Class Codes: G05D 1/0253 extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow determining position or orientation from images G06T7/70
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/30244 Camera pose
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/77 using statistical methods
• CPC Class Codes: G06V 20/10 Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59

Abstract

A simultaneous localization and map building method of a mobile robot including an omni-directional camera. The method includes acquiring an omni-directional image from the omni-directional camera, dividing the obtained omni-directional image into upper and lower images according to a preset reference to generate a first image, which is the lower image, and a second image, which is the upper image, extracting feature points from the first image and calculating visual odometry information calculating visual odometry information to track locations of the extracted feature points based on a location of the omni-directional camera, and performing localization and map building of the mobile robot using the calculated visual odometry information and the second image as an input of an extended Kalman filter.

Claims

1. A simultaneous localization and map building method of a mobile robot including an omni-directional camera, comprising:

acquiring an omni-directional image from the omni-directional camera;

dividing the acquired omni-directional image according to a preset reference to generate a lower image and an upper image;

extracting feature points from the lower image and calculating visual odometry information to track locations of the extracted feature points based on a location of the omni-directional camera; and

performing localization and map building of the mobile robot using the calculated visual odometry information and the upper image as an input of an extended Kalman filter.

2. The simultaneous localization and map building method of claim 1, wherein the omni-directional camera is installed in parallel to a horizontal plane at a front surface of the mobile robot.

3. The simultaneous localization and map building method of claim 1, wherein said calculating visual odometry information includes:

calculating locations of the extracted feature points based on a current location of the omni-directional camera;

calculating a rotation vector and a translation vector according to movement of the omni-directional camera; and

tracking locations of the extracted feature points by calculating locations of the extracted feature points based on a moved location of the omni-directional camera according to the calculated rotation vector and translation vector.

4. The simultaneous localization and map building method of claim 3, wherein said extracting feature points from the lower image includes extracting new feature points from an area of the lower image excluding an area in which the location-tracked feature points are present.

5. The simultaneous localization and map building method of claim 1, wherein the preset reference is a height of the omni-directional camera based on the ground.

6. The simultaneous localization and map building method of claim 1, wherein said performing localization and map building of the mobile robot includes:

extracting feature points from the upper image;

predicting locations of the robot and the extracted feature points of the upper image using the calculated visual odometry information as a traveling input; and

updating the locations of the robot and the extracted feature points of the upper image by updating the locations of the robot and the extracted feature points of the upper image based on the upper image.

7. The simultaneous localization and map building method of claim 1, wherein at least one of said dividing, said extracting, said calculating and said performing are performed by a computer.

8. A mobile robot including an omni-directional camera, comprising:

an image processor to divide an omni-directional image captured through the omni-directional camera according to a preset reference to generate a lower image and an upper image;

a visual odometry information calculator to extract feature points from the lower image and track locations of the extracted feature points based on the omni-directional camera; and

a simultaneous location and map building (SLAM) executor to perform localization and map building of the mobile robot using visual odometry information calculated by the visual odometry information calculator and the upper image as an input of an extended Kalman filter.

9. The mobile robot of claim 8, wherein the omni-directional camera is installed in parallel to a horizontal plane at a front surface of the mobile robot.

10. The mobile robot of claim 8, wherein the visual odometry information calculator:

calculates locations of the extracted feature points based on a current location of the omni-directional camera;

calculates a rotation vector and a translation vector according to movement of the omni-directional camera; and

tracking the locations of the extracted feature points by recalculating the locations of the extracted feature points based on a moved location of the omni-directional camera according to the calculated rotation vector and translation vector.

11. The mobile robot of claim 10, wherein the visual odometry information calculator extracts new feature points from an area of the lower image excluding an area in which the location-tracked feature points are present.

12. The mobile robot of claim 8, wherein the image processor divides the captured omni-directional image according to the preset reference which is a height of the omni-directional camera based on the ground.

13. The mobile robot of claim 8, wherein the SLAM executor:

extracts feature points from the upper image;

predicts locations of the mobile robot and the extracted feature points of the upper image using the calculated visual odometry information as a traveling input; and

updates the locations of the mobile robot and the extracted feature points of the upper image by updating the locations of the robot and the extracted feature points of the upper image based on the second image.

14. A method comprising:

dividing, by a computer, an omni-directional image acquired from an omni-directional camera of a mobile robot according to a preset reference to generate a lower image and an upper image;

calculating, by a computer and based on a location of the omni-directional camera, visual odometry information to track locations of feature points extracted from the lower image; and

performing, by a computer, localization and map building of the mobile robot via an extended Kalman filter using the calculated visual odometry information and the upper image.