| CPC G06V 10/80 (2022.01) [A47L 9/2826 (2013.01); A47L 9/2852 (2013.01); G01C 21/206 (2013.01); G06V 10/30 (2022.01); G06V 20/58 (2022.01); A47L 2201/04 (2013.01)] | 16 Claims |
|
1. A method for controlling a robot cleaner comprising:
a map data collecting step of collecting first information acquired when a cleaning target area is photographed by a depth camera and second information acquired when a cleaning target area is photographed by an infrared (IR) sensor;
a depth data filtering step of removing, from the first information, information determined to be noise when detecting obstacles; and
a data restoration step of determining whether there is a lost part in information about obstacles by comparing the second information with the information that was determined to be noise and removed in the depth data filtering step, and restoring the lost part when the lost part is present in the information about obstacles,
wherein the depth camera and the IR sensor simultaneously collect information about the target area while the robot cleaner moves, and
wherein the depth data filtering step includes:
a distance comparison step in which, if a difference between an actual position and a reference position corresponding to a distance value where each pixel of the depth camera is detected is a predetermined distance or more, the corresponding information is determined to be noise and removed.
|
|
15. A method for controlling a robot cleaner comprising:
a map data collecting step of collecting first information acquired when a cleaning target area is photographed by a depth camera and second information acquired when a cleaning target area is photographed by an infrared (IR) sensor;
a depth data filtering step of removing, from the first information, information determined to be noise when detecting obstacles; and
a data restoration step of determining whether there is a lost part in information about obstacles by comparing the second information with the information that was determined to be noise and removed in the depth data filtering step, and restoring the lost part when the lost part is present in the information about obstacles,
wherein the depth camera and the IR sensor simultaneously collect information about the target area while the robot cleaner moves, and
wherein the second information is acquired by:
extracting feature points from an image photographed by the IR sensor;
determining corners by interconnecting the feature points; and
determining information about obstacles by interconnecting corners when the presence of the corners is decided.
|
|
16. A method for controlling a robot cleaner comprising:
a map data collecting step of collecting first information acquired when a cleaning target area is photographed by a depth camera and second information acquired when a cleaning target area is photographed by an infrared (IR) sensor;
a depth data filtering step of removing, from the first information, information determined to be noise when detecting obstacles; and
a data restoration step of determining whether there is a lost part in information about obstacles by comparing the second information with the information that was determined to be noise and removed in the depth data filtering step, and restoring the lost part when the lost part is present in the information about obstacles,
wherein the depth camera and the IR sensor simultaneously collect information about the target area while the robot cleaner moves, and
wherein the data restoration step includes:
when information determined to be noise is removed from the first information, determining whether a discontinuous part is present in the information determined to be noise, and when the presence of the discontinuous part is determined, determining the presence of a lost part in the information determined to be noise.
|