US 11,911,919 B2
Method and computing system for performing grip region detection
Jose Jeronimo Moreira Rodrigues, Tokyo (JP); Rosen Nikolaev Diankov, Tokyo (JP); Jinze Yu, Tokyo (JP); Ahmed Abouelela, Tokyo (JP); and Yoshiki Kanemoto, Tokyo (JP)
Assigned to MUJIN, INC., Tokyo (JP)
Filed by MUJIN, INC., Tokyo (JP)
Filed on Mar. 5, 2021, as Appl. No. 17/193,268.
Prior Publication US 2022/0281114 A1, Sep. 8, 2022
Int. Cl. B25J 9/16 (2006.01)
CPC B25J 9/1697 (2013.01) 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
a communication interface configured to communicate with a robot having an end effector apparatus, and with a camera having a camera field of view;
at least one processing circuit configured, when a flexible object is in the camera field of view, to:
receive image information that represents an object surface associated with the flexible object, wherein the image information is generated by the camera;
identify, as a grip region, a surface region of the object surface that satisfies a defined smoothness condition and has a region size that is larger than or equal to a defined region size threshold, wherein the grip region is identified based on the image information;
identify, as a safety region, a three-dimensional (3D) region indicative of a sag portion of the flexible object outside of the grip region in one or more horizontal dimensions, and which extends from the one or more horizontal dimensions along a vertical dimension that is perpendicular to the one or more horizontal dimensions;
perform robot motion planning based on the grip region and the safety region, wherein the robot motion planning is for:
gripping of the flexible object at the grip region by the end effector apparatus, and
avoiding other objects in an environment with the robot; and
communicate commands to the robot to cause the robot to perform the planned motion.