Localization method for a moving robot

• US 20080232678A1
• Filed: 01/18/2008
• Published: 09/25/2008
• Est. Priority Date: 03/20/2007
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Roh, Kyung Shik, Min, Seung Ki, Han, Woosup, Yoon, Sukjune
• Granted Patent: US 8,588,512 B2
• Time in Patent Office: Days
• US Class Current: 382/153
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle

Abstract

A localization method of a moving robot is disclosed in which the moving robot includes: capturing a first omni-directional image by the moving robot; confirming at least one node at which a second omni-directional image having a high correlation with the first omni-directional image is captured; and determining that the moving robot is located at the first node when the moving robot reaches a first node, at which a second omni-directional image having a highest correlation with the first omni-directional image is captured, from among the at least one node.

Claims

1. A localization method of a moving robot, the method comprising:

capturing a first omni-directional image by the moving robot;

confirming at least one node at which a second omni-directional image having a high correlation with the first omni-directional image is captured; and

determining that the moving robot is located at the first node when the moving robot reaches a first node, at which a second omni-directional image having a highest correlation with the first omni-directional image is captured, from among the at least one node.

2. The method according to claim 1, further comprising:

determining that the moving robot reaches the first node when a correlation coefficient between the second omni-directional image of the first node and the first captured omni-directional image is higher than a reference value.

3. The method according to claim 1, further comprising:

determining that the moving robot reaches the first node when the current first node is not changed from a previous first node and the correlation coefficient between the second omni-directional image of the first node and the first captured omni-directional image gradually increases and then decreases.

4. The method according to claim 1, further comprising:

moving the moving robot to continuously move to the first node when the correlation coefficient between the second omni-directional image of the first node and the first captured omni-directional image gradually increases and then does not decrease.

5. The method according to claim 3, further comprising:

moving the moving robot to move to a new first node when the first node is changed from a previous first node.

6. The method according to claim 4, further comprising:

determining, by the moving robot, whether the remaining nodes other than the first node from among the at least one node are effective or not when the moving robot moves to the first node, wherein

the moving direction of the moving robot is maintained when the number of effective nodes is set to “

0”

, the moving robot moves from the effective node to the first node when the number of effective nodes is set to “

1”

, and the moving robot moves from a gravity center of a triangle composed of the first node and two effective nodes to the first node when the number of effective nodes is set to “

2”

.

7. The method according to claim 1, wherein each of the correlations is indicative of a correlation between current lines of two omni-directional images.

8. The method according to claim 1, wherein each of the correlations is represented by a correlation coefficient, and the correlation coefficient is calculated by the following equation:

$\begin{array}{cc}\mathrm{\rho <br><br>}<br><br>\left(\mathrm{\tau <br><br>}\right)=\frac{C_{\mathrm{xy}}<br><br>\left(\mathrm{\tau <br><br>}\right)}{\sqrt{C_{\mathrm{xx}}<br><br>\left(0\right)·<br><br>C_{\mathrm{yy}}<br><br>\left(0\right)}}& \left[\mathrm{Equation}\right]\end{array}$

where ρ

(τ

) is indicative of a correlation coefficient, τ

is indicative of a rotation angle, C_xy is indicative of a cross-correlation value, and C_xx and C_yy are indicative of auto-correlation values.

9. The method according to claim 8, wherein the cross-correlation value is calculated by a Fast Fourier Transform (FFT).

16. The method according to claim 5, further comprising:

determining, by the moving robot, whether the remaining nodes other than the new first node from among the at least one node are effective or not when the moving robot moves to the new first node, wherein

the moving direction of the moving robot is maintained when the number of effective nodes is set to “

0”

, the moving robot moves from the effective node to the new first node when the number of effective nodes is set to “

1”

, and the moving robot moves from a gravity center of a triangle composed of the new first node and two effective nodes to the new first node when the number of effective nodes is set to “

2”

.

10. A localization method of a moving robot, the method comprising:

capturing a first omni-directional image by the moving robot;

confirming at least one node at which a second omni-directional image having a high correlation with the first omni-directional image is captured; and

determining that the moving robot is located at the first node when the moving robot reaches a first node, at which a second omni-direction image having a highest correlation with the first captured omni-directional image is captured, and moving the moving robot to move to the first node when the moving robot does not reach the first node.

11. The method according to claim 10, further comprising:

determining whether the remaining nodes other than the first node from among the at least one node are effective or not, and moving the moving robot to move to a heading angle determined according to the number of effective nodes, so that the moving robot approaches the first node.

12. The method according to claim 11, wherein:

the number of the at least one node is “

3”

, the moving direction of the moving robot is set to the heading angle when the number of effective nodes is “

0”

, the direction from the effective node to the first node is set to the heading angle direction when the number of effective nodes is “

1”

, and the direction from a gravity center of a triangle composed of the first node and two effective nodes to the first node is set to the heading angle when the number of effective nodes is “

2”

.

13. The method according to claim 10, wherein each of the correlations is indicative of a correlation between current lines of two omni-directional images.

14. The method according to claim 10, wherein each of the correlations is represented by a correlation coefficient, and the correlation coefficient is calculated by the following equation:

$\begin{array}{cc}\mathrm{\rho <br><br>}<br><br>\left(\mathrm{\tau <br><br>}\right)=\frac{C_{\mathrm{xy}}<br><br>\left(\mathrm{\tau <br><br>}\right)}{\sqrt{C_{\mathrm{xx}}<br><br>\left(0\right)·<br><br>C_{\mathrm{yy}}<br><br>\left(0\right)}}& \left[\mathrm{Equation}\right]\end{array}$

where ρ

(τ

) is indicative of a correlation coefficient, τ

is indicative of a rotation angle, C_xy is indicative of a cross-correlation value, and C_xx and C_yy are indicative of auto-correlation values.

15. The method according to claim 14, wherein the cross-correlation value is calculated by a Fast Fourier Transform (FFT).

17. The method according to claim 11, wherein the nodes are determined to be effective when the remaining nodes other than the first node are within an effective range of the first node.

18. A moving robot, comprising:

an omni-directional camera capturing an omni-directional image by the moving robot;

a correlation coefficient calculator calculating a correlation coefficient between the captured omni-directional image and at least a second omni-directional image captured at least at one node location of a movable place of the moving robot, the at least one node location being the node location having the highest correlation coefficient from among a plurality of node locations; and

a controller moving the moving robot to the first node location when the correlation coefficient calculator calculates that the correlation coefficient is not greater than a reference value and the current first node location is not equal to a previous first node location and determining that the moving robot is at the first node location when the correlation coefficient is greater than the reference value.

19. The moving robot according to claim 18, wherein the controller moves the moving robot to the first node location when the current first node location is not equal to a previous first node location.

20. The moving robot according to claim 18, wherein the controller determines that the moving robot is at the first node location when the correlation coefficient is not greater than the reference value, the current first node location is equal to the previous first node location, and the correlation coefficient of the current first node location is less than a correlation coefficient of the previous first node location, the correlation coefficient of the previous first node location being greater than correlation coefficients of other first node locations prior to the previous first node location.

21. The moving robot according to claim 18, wherein the controller moves the moving robot to the first node location when the correlation coefficient is not greater than the reference value, the current first node location is equal to the previous first node location, and the correlation coefficient of the current first node location is not less than a correlation coefficient of the previous first node location.