Mobile robot control method

1. A method of controlling a mobile robot, the method comprising:

acquiring an image of an inside of a traveling zone as the mobile robot travels in the traveling zone;

performing a point-based feature point extraction by extracting a first feature point from the acquired image;

performing a block-based feature point extraction by dividing the acquired image into blocks having a predetermined size and extracting a second feature point from each of the divided block-unit images;

performing a point-based feature point matching by comparing feature points of a first node corresponding to a current location of the mobile robot and a second node located within a predetermined reference distance from the first node using the first feature point;

performing a block-based feature point matching by matching the feature points of the first node and the second node using the second feature point;

determining the current location based on the point-based feature point matching and the block-based feature point matching; and

storing the determined current location in association with the first feature point and the second feature point in a map.

2. The method according to claim 1 , wherein the point-based feature point extraction comprises creating a descriptor corresponding to the extracted first feature point based on distribution characteristics of a brightness gradient of pixels belonging to a certain area around the extracted first feature point.

3. The method according to claim 1 , wherein the block-based feature point extraction comprises creating a descriptor by block based on distribution characteristics of a brightness gradient of the block-unit images.

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

acquiring traveling information, including an amount of movement of the mobile robot; and

acquiring the image using an image acquisition unit when the amount of movement from a previous node is greater than a threshold value.

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

acquiring traveling information, including an amount of movement of the mobile robot; and

selecting a key frame image from among images acquired by an image acquisition unit, the key frame image corresponding to an amount of movement from a previous node greater than a threshold value; and

performing the point-based feature point extraction and the block-based feature point extraction using the key frame image.

6. A method of controlling a mobile robot, the method comprising:

acquiring an image of an inside of a traveling zone as the mobile robot travels in the traveling zone;

performing block-based feature point extraction by dividing the acquired image into blocks having a predetermined size and extracting first feature points from each of the divided block-unit images;

performing block-based feature point matching by comparing the extracted first feature points with second feature points stored in a map, the second feature points having been extracted using the block-based feature point extraction; and

determining a current location of the mobile robot based on a result of the block-based feature point matching.

7. The method according to claim 6 , wherein the block-based feature point extraction comprises creating a descriptor by block based on distribution characteristics of a brightness gradient of the block-unit images.

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

performing a point-based feature point extraction by extracting a feature point from the acquired image; and

matching the extracted feature point with one or more feature points stored on the map, the stored feature points having been obtained using the point-based feature point extraction; and

determining the current location of the mobile robot based on a first result of the point-based feature point matching and a second result of the block-based feature point matching.

9. The method according to claim 8 , wherein the point-based feature point extraction comprises creating a descriptor corresponding to the extracted feature point based on distribution characteristics of a brightness gradient of pixels belonging to a certain area around the extracted feature point.

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

acquiring traveling information, including an amount of movement of the mobile robot; and

acquiring the image using an image acquisition unit when the amount of movement from a previous node is greater than a threshold value.

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

acquiring traveling information, including an amount of movement of the mobile robot; and

selecting a key frame image from images acquired by an image acquisition unit, the key frame image corresponding to the amount of movement from a previous node greater than a threshold value; and

performing the block-based feature point extraction on the key frame image.

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

acquiring illuminance information through a sensor unit; and

performing the block-based feature point extraction when the illuminance information corresponds to a low-illuminance environment.

13. The method according to claim 12 , further comprising:

performing point-based feature point extraction by extracting a feature point from the acquired image when the illuminance information does not correspond to the low-illuminance environment;

performing a point-based feature point matching by comparing the extracted feature point with one or more feature points stored on the map. the stored feature points having been obtained using the point-based feature point extraction; and

determining the current location of the mobile robot based on the point-based feature point matching.

14. The method according to claim 13 , wherein the point-based feature point extraction comprises creating a descriptor corresponding to the extracted feature point based on distribution characteristics of a brightness gradient of pixels belonging to a certain area around the extracted feature point.

15. The method according to claim 6 , wherein determining the current location comprises:

setting a plurality of particles, which are location candidates, based on the block-based feature point matching;

determining weights for the particles;

determining the current location based on the weights; and

resampling the particles.