Method and apparatus for simultaneous localization and mapping of robot

• US 20100094460A1
• Filed: 07/22/2009
• Published: 04/15/2010
• Est. Priority Date: 10/09/2008
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Hyoung-ki, Choi, Ki-wan, Park, Ji-young
• Granted Patent: US 8,855,819 B2
• Time in Patent Office: Days
• US Class Current: 700/251
• CPC Class Codes: B25J 9/1692 Calibration of manipulator
• CPC Class Codes: G05B 2219/37563 Ccd, tv camera
• CPC Class Codes: G05B 2219/39388 Visual compliance, xy constraint is 2-D image, z position controlled
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/30244 Camera pose
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle

Abstract

A SLAM of a robot is provided. The position of a robot and the position of feature data may be estimated by acquiring an image of the robot'"'"'s surroundings, extracting feature data from the image, and matching the extracted feature data with registered feature data. Furthermore, measurement update is performed in a camera coordinate system and an appropriate assumption is added upon coordinate conversion, thereby reducing non-linear components and thus improving the SLAM performance.

Claims

1. A simultaneous localization and mapping (SLAM) apparatus of a robot, comprising:

a converter to convert a two-dimensional (2D) position of measured feature data to a three-dimensional (3D) position; and

an updater to update registered feature data by comparing the converted 3D position of the measured feature data with a 3D position of the registered feature data.

2. The SLAM apparatus of claim 1, wherein the converter converts the 2D position of the measured feature data to a 3D position in a camera coordinate system of a camera for acquiring the feature data.

3. The SLAM apparatus of claim 1, wherein the updater updates the registered feature data in a camera coordinate system of a camera for acquiring the feature data.

4. The SLAM apparatus of claim 1, wherein the feature data is a feature point.

5. The SLAM apparatus of claim 4, wherein, in the converting of the 2D position to the 3D position, the converter sets a depth of the feature point to a predetermined value and sets an error covariance in the depth direction to infinity.

6. The SLAM apparatus of claim 1, wherein the feature data is a feature line, and a position of the feature line is defined to be a 2D position of both defined end points of the feature line.

7. The SLAM apparatus of claim 6, wherein, in the converting of the 2D position to the 3D position, the converter sets a depth of each end point to a predetermined value and sets an error covariance in the depth direction and an error covariance in a direction parallel to the feature line to infinity.

8. A SLAM method of a robot, comprising:

converting a two-dimensional (2D) position of measured feature data to a three-dimensional (3D) position; and

updating registered feature data by comparing the converted 3D position of the measured feature data with a 3D position of the registered feature data.

9. The SLAM method of claim 8, wherein the 2D position of the measured feature data is converted to a 3D position in a camera coordinate system of a camera for acquiring the feature data.

10. The SLAM method of claim 8, wherein updating is performed in a camera coordinate system of a camera for acquiring the feature data.

11. The SLAM method of claim 8, wherein the feature data is a feature point.

12. The SLAM method of claim 11, wherein the converting further comprises setting a depth of the feature point to a predetermined value and setting an error covariance in the depth direction to infinity when converting the 2D position to the 3D position.

13. The SLAM method of claim 8, wherein the feature data is a feature line, and a position of the feature line is defined to be a 2D position of both defined end points of the feature line.

14. The SLAM method of claim 13, wherein the converting further comprises setting a depth of each end point to a predetermined value and setting an error covariance in the depth direction and an error covariance in a direction parallel to the feature line to infinity when converting the 2D position to the 3D position.