SYSTEMS AND METHODS FOR PRECISELY ESTIMATING A ROBOTIC FOOTPRINT FOR EXECUTION OF NEAR-COLLISION MOTIONS
Systems and methods for precisely estimating a robotic footprint for execution of near-collision motions are disclosed herein. According to at least one non-limiting exemplary embodiment, a robot may switch from using a computer readable map to a sensor which senses at least a portion of the robot to navigate close by objects.
1 . A method for maneuvering a robot, comprising:
navigating the robot using a computer readable map, the computer readable map comprising a footprint of the robot;
receiving sensor data from a sensor coupled to the robot, the sensor data comprises a field of view which encompasses at least a portion of a body of the robot;
detecting the footprint of the robot within a threshold distance to one or more objects localized on the computer readable map;
detecting a portion of the sensor data which senses the portion of the body of the robot within the field of view;
determining a distance between the body of the robot and the one or more objects based on the portions of the sensor data which senses the body of the robot and the one or more objects;
navigating the robot until the distance is below a threshold value; and
stopping the robot.
2 . The method of claim 1 , wherein,
the sensor comprises a depth camera;
the sensor data corresponds to depth imagery; and
the portion of the sensor data which senses the portion of the body of the robot corresponds to pixels of the depth imagery.
3 . The method of claim 2 , further comprising:
producing a pixel mask, the pixel mask corresponding to pixels of the depth imagery which depict the portion of the body of the robot within the field of view; and
updating the pixel mask based on the receipt of at least one additional depth image.
4 . The method of claim 2 , wherein,
the distance between the body of the robot and the one or more objects is measured based on depth values of the depth imagery.
5 . The method of claim 3 , wherein,
the portion of the body of the robot is detected within the depth imagery using at least one of: (i) motion analysis between two or more successive depth images, wherein the portion of the body of the robot is stationary; (ii) pixel color analysis between two or more depth images, wherein pixels comprising a large color differential between the two or more images are determined to not correspond to the robot; or (iii) expected distances between the depth camera and the robot based on calibration values for the depth camera.
6 . The method of claim 1 , further comprising
requesting human assistance using communications units coupled to the robot, the request for assistance comprising the robot to perform at least one of: (i) emitting an auditory noise; (ii) emitting a visual indication; or (iii) transmitting a signal using a cellular or Wi-Fi network to a device of a human.
7 . A non-transitory computer readable storage medium comprising a plurality of computer readable instructions embodied thereon which, when executed by at least one controller, causes the at least one controller to:
navigate a robot using a computer readable map, the computer readable map comprising a footprint of the robot;
receive sensor data from a sensor coupled to the robot, the sensor data comprises a field of view which encompasses at least a portion of a body of the robot;
detect the robot footprint is within a threshold distance to one or more objects localized on the computer readable map;
detect a portion of the sensor data which senses the portion of the body of the robot within the field of view;
determine a distance between the body of the robot and the one or more objects based on the portions of the sensor data which senses the body of the robot body and the one or more objects;
navigate the robot until the distance is below a threshold value; and
stop the robot.
8 . The non-transitory computer readable storage medium of claim 7 , wherein,
the sensor comprises a depth camera;
the sensor data corresponds to depth imagery; and
the portion of the sensor data which senses the portion of the body of the robot corresponds to pixels of the depth imagery.
9 . The non-transitory computer readable storage medium of claim 8 , wherein the at least one controller is further configured to execute the computer readable instructions to:
produce a pixel mask, the pixel mask corresponding to pixels of the depth imagery which depict the portion of the body of the robot within the field of view; and
updating the pixel mask based on the receipt of at least one additional depth image.
10 . The non-transitory computer readable storage medium of claim 8 , wherein, the distance between the body of the robot and the one or more objects is measured based on depth values of the depth imagery.
11 . The non-transitory computer readable storage medium of claim 9 , wherein,
the portion of the body of the robot is detected within the depth imagery using at least one of: (i) motion analysis between two or more successive depth images, wherein the portion of the body of the robot is stationary; (ii) pixel color analysis between two or more depth images, wherein pixels comprising a large color differential between the two or more images are determined to not correspond to the robot; or (iii) expected distances between the depth camera and the robot based on calibration values for the depth camera.
12 . The non-transitory computer readable storage medium of claim 7 , wherein the at least one controller is further configured to execute the computer readable instructions to:
request human assistance using communications units coupled to the robot, the request for assistance comprising the robot to perform at least one of: (i) emitting an auditory noise; (ii) emitting a visual indication; or (iii) transmitting a signal using a cellular or Wi-Fi network to a device of a human.
13 . A robotic system, comprising:
a memory comprising a plurality of computer readable instructions; and
at least one controller configured to execute the computer readable instructions to:
navigate the robotic system using a computer readable map, the computer readable map comprises a footprint of the robotic system;
receive sensor data from a sensor coupled to the robotic system, the sensor data comprises a field of view which encompasses at least a portion of a body of the robotic system;
detect the robotic system footprint within a threshold distance to one or more objects localized on the computer readable map;
detect a portion of the sensor data which senses the portion of the body of the robotic system within the field of view;
determine a distance between the body of the robotic system and the one or more objects based on the portions of the sensor data which senses the body of the robotic system and the one or more objects;
navigate the robotic system until the distance is below a threshold value; and
stop the robotic system.
14 . The robotic system of claim 13 , wherein,
the sensor comprises a depth camera;
the sensor data corresponds to depth imagery; and
the portion of the sensor data which senses the portion of the body of the robotic system corresponds to pixels of the depth imagery.
15 . The robotic system of claim 14 , wherein the at least one controller is further configured to execute the computer readable instructions to:
produce a pixel mask, the pixel mask corresponding to pixels of the depth imagery which depict the portion of the body of the robotic system within the field of view; and
update the pixel mask based on the receipt of at least one additional depth image.
16 . The non-transitory computer readable storage medium of claim 14 , wherein,
the distance between the body of the robotic system body and the one or more objects is measured based on depth values of the depth imagery.
17 . The non-transitory computer readable storage medium of claim 15 , wherein,
the portion of the body of the robotic system is detected within the depth imagery using at least one of: (i) motion analysis between two or more successive depth images, wherein the portion of the body of the robotic system is stationary; (ii) pixel color analysis between two or more depth images, wherein pixels comprising a large color differential between the two or more images are determined to not correspond to the robotic system; or (iii) expected distances between the depth camera and the robotic system based on calibration values for the depth camera.
18 . The non-transitory computer readable storage medium of claim 13 , wherein the at least one controller is further configured to execute the computer readable instructions to:
request human assistance using communications units of the robotic system, the request for assistance comprises the robotic system to perform at least one of: (i) emitting an auditory noise; (ii) emitting a visual indication; or (iii) transmitting a signal using a cellular or Wi-Fi network to a device of a human.
19 . The robotic system of claim 13 , wherein,
the robotic system is a floor cleaning robot.