Method of updating map in fusion SLAM and robot implementing same

1. A robot configured to update a map in fusion simultaneous localization and mapping (SLAM), the robot comprising;

a moving unit configured to move the robot;

a map storage configured to store a map of an area in which the robot moves,

wherein the map has a double-layer structure, a first layer of the double-layer structure being a backbone and a second layer of the double-layer structure including a light detection and ranging (LiDAR) branch and a visual branch,

wherein the backbone is information on a trajectory of the robot and includes one or more frame nodes corresponding to the trajectory,

wherein the LiDAR branch comprises LiDAR frames with a LiDAR sensing value,

wherein the visual branch comprises visual frames with a camera sensing value, and

wherein at least one of the LiDAR frames is connected to the one or more frame nodes of the backbone, and at least one of the visual frames is connected to the one or more frame nodes of the backbone;

a first sensor configured to acquire first type information required for the robot to perform SLAM;

a second sensor configured to acquire second type information required for the robot to perform SLAM; and

a controller configured to estimate a current position of the robot using a second map of a second type, to update a first map using the first type information with respect to the estimated position, and to store the updated first map in the map storage,

wherein the controller comprises an artificial intelligent unit, and the controller is further configured to acquire information on an area with high accuracy of localization or an area with low accuracy of localization and provide the acquired information to the artificial intelligent unit;

wherein the artificial intelligent unit is configured to compare the provided acquired information on the area with high accuracy of localization or the area with low accuracy of localization with information previously stored in the map storage, and determine whether to update the map,

wherein when the LiDAR branch constitutes the first map, then the visual branch constitutes the second map, and

wherein when the visual branch constitutes the first map, then the LiDAR branch constitutes the second map.

2. The robot of claim 1 , wherein the first sensor is a LiDAR sensor and the second sensor is a camera sensor,

the map storage stores a visual map that is the second map,

the controller loads the visual map to estimate the current position of the robot, and

the controller adds information, sensed by the LiDAR sensor with respect to the estimated position, to the first map.

3. The robot of claim 2 , wherein the first map is any one of a LiDAR map stored in the map storage or a computer aided design (CAD) map comprised of CAD information.

4. The robot of claim 3 , wherein the CAD map is map information in relation to building design.

5. The robot of claim 2 , wherein the controller maintains information that constitutes the visual map to maintain an alignment between the visual map and the first map.

6. The robot of claim 1 , wherein the first sensor is a camera sensor, and the second sensor is a light detection and ranging (LiDAR) sensor,

the map storage stores a LiDAR map that is the second map,

the controller loads the LiDAR map to estimate the current position of the robot, and

the controller adds information, sensed by the camera sensor with respect to the estimated position, to the first map.

7. The robot of claim 6 , wherein the first map is any one of a visual map stored in the map storage or a computer aided design (CAD) map comprised of CAD information.

8. The robot of claim 6 , wherein the controller maintains information that constitutes the LiDAR map to maintain an alignment between the LiDAR map and the first map.

9. The robot of claim 1 , wherein the controller maintains the second map and optimizes pieces of added information of the first map.

10. The robot of claim 9 , wherein the controller does not change a coordinate of a point of interest (POI) while maintaining the second map and optimizing the pieces of added information of the first map.

11. A method of updating a map in fusion simultaneous localization and mapping (SLAM), the method comprising:

controlling, by a controller of a robot, the robot to move;

based on movement of the robot, storing, by the controller of the robot, a map having a double-layer structure, a first layer of the double-layer structure being a backbone and a second layer of the double-layer structure including a light detection and ranging (LiDAR) branch and a visual branch,

wherein the backbone is information on a trajectory of the robot and includes one or more frame nodes corresponding to the trajectory,

wherein the LiDAR branch comprises LiDAR frames with a LIDAR sensing value,

wherein the visual branch comprises visual frames with a camera sensing value, and

wherein at least one of the LiDAR frames is connected to the one or more frame nodes of the backbone, and at least one of the visual frames is connected to the one or more frame nodes of the backbone;

acquiring, by a first sensor of the robot, first type information;

acquiring, by a second sensor of the robot, second type information;

estimating, by the controller of the robot, a current position of the robot using a second map of a second type stored in a map storage of the robot;

updating, by the controller, a first map using the first type information with respect to the estimated position; and

storing, by the controller, the updated first map in the map storage,

wherein the controller comprises an artificial intelligent unit, and the controller is configured to acquire information on an area with high accuracy of localization or an area with low accuracy of localization and provide the acquired information to the artificial intelligent unit;

wherein the artificial intelligent unit is configured to compare the provided acquired information on the area with high accuracy of localization or the area with low accuracy of localization with information previously stored in the map storage, and determine whether to update the map;

wherein when the LiDAR branch constitutes the first map, then the visual branch constitutes the second map; and

wherein when the visual branch constitutes the first map, then the LiDAR branch constitutes the second map.

12. The method of claim 11 , wherein the first sensor is a LiDAR sensor and the second sensor is a camera sensor,

wherein the map storage stores a visual map that is the second map, and

wherein the method further comprises:

loading, by the controller, the visual map to estimate the current position of the robot; and

adding, by the controller, information, sensed by the LiDAR sensor with respect to the estimated position, to the first map.

13. The method of claim 12 , wherein the first map is any one of a LiDAR map stored in the map storage or a computer aided design (CAD) map comprised of CAD information.

14. The method of claim 13 , wherein the CAD map is map information in relation to building design.

15. The method of claim 12 , further comprising:

maintaining, by the controller, information that constitutes the visual map to maintain an alignment between the visual map and the first map.

16. The method of claim 9 , wherein the first sensor is a camera sensor and the second sensor is a light detection and ranging (LiDAR) sensor,

wherein the map storage stores a LiDAR map that is the second map, and

wherein the method further comprises:

loading, by the controller, the LiDAR map to estimate the current position of the robot; and

adding, by the controller, information, sensed by the camera sensor with respect to the estimated position, to the first map.

17. The method of claim 16 , wherein the first map is any one of a visual map stored in the map storage or a computer aided design (CAD) map comprised of CAD information.

18. The method of claim 16 , further comprising:

maintaining, by the controller, information that constitutes the LiDAR map to maintain an alignment between the LiDAR map and the first map.

19. The method of claim 11 , further comprising:

maintaining, by the controller, the second map and optimizing pieces of added information of the first map.

20. The method of claim 19 , wherein the controller does not change a coordinate of a point of interest (POI) while maintaining the second map and optimizing the pieces of added information of the first map.