CPC B25J 9/1687 (2013.01) [B25J 9/0093 (2013.01); B25J 9/161 (2013.01); B25J 9/1653 (2013.01); B25J 9/1664 (2013.01); B25J 9/1697 (2013.01); B25J 11/0045 (2013.01); B25J 13/06 (2013.01)] | 16 Claims |
1. A method comprising:
based on a first set of depth-image data sampled during a first time period with a camera oriented towards a foodstuff bin, generating a representation of a three-dimensional surface of bulk foodstuff within the foodstuff bin;
based on the representation, executing a first pick with a food utensil of a robot;
maintaining the representation over a second time period after execution of the first pick, comprising:
updating a portion of the representation of the three-dimensional surface based on a second set of data, wherein a remainder of the representation is based on the first set of depth-image data; and
contemporaneously with maintaining the representation of the three— dimensional surface, repeatedly:
based on a target portion amount of the bulk foodstuff, generating a pick depth map for the three-dimensional surface, using the representation;
based on the pick depth map, selecting a pick target associated with a coordinate position and a pick-depth below the three-dimensional surface of the foodstuff;
determining control instructions for the robot based on the pick target, the control instructions comprising a robot trajectory; and
executing a pick with the food utensil of the robot based on the control instructions.
|