Back to results
US 12,252,355 B2Granted
Robotic singulation system sensor
AssigneeDexterity, Inc.
InventorsTimothy Ryan, Zhouwen Sun, Harry Zhe Su, Talbot Morris-Downing, David Leo Tondreau, III, Toby Leonard Baker, Ayush Sharma, Raphael Georg Wirth
Life of this patent
Estimating the term…
Abstract
A robotic system comprising an end effector-mounted sensor is disclosed. In various embodiments, a robotic arm is manipulated to move a sensor to a position such that an object of interest is within a read range of the sensor. A sensor data read by the sensor is received via a communication interface. The sensor data is used to determine an attribute of the object; and the determined attribute of the object is used to determine a plan to grasp and move the object.
Claims13
A system, comprising:
- a communication interface;
- and
- a processor coupled to the communication interface and configured to:
- receive from a computer vision system video segmentation data indicating an identified object of interest in a workspace;
- manipulate a robotic arm to move a first sensor to a position such that the identified object of interest is within a read range of the sensor;
- receive via the communication interface a first sensor data read by the first sensor, the first sensor data comprising a first data element associated with a first object and a second data element associated with a second object;
- receive via the communication interface a second sensor data read by a second sensor positioned statically in the workspace, the second sensor data comprising the second data element associated with the second object but not the first data element associated with the first object;
- determine based on first sensor data and the second sensor data that the identified object comprises both the first object and the second object;
- update a view of the workspace, based at least in part on said determination, to determine an updated view in which the first object and the second object are separately identified;
- use the first data element to determine an attribute of the first object;
- and
- use the determined attribute of the first object and the updated view to determine a plan to grasp and move the first object.
9 more elements
A method, comprising:
- receiving from a computer vision system video segmentation data indicating an identified object of interest in a workspace;
- manipulating a robotic arm to move a first sensor to a position such that the identified object of interest is within a read range of the sensor;
- receiving via a communication interface a first sensor data read by the first sensor, the first sensor data comprising a first data element associated with a first object and a second data element associated with a second object;
- receiving via the communication interface a second sensor data read by a second sensor positioned statically in the workspace, the second sensor data comprising the second data element associated with the second object but not the first data element associated with the first object;
- determining based on first sensor data and the second sensor data that the identified object comprises both the first object and the second object;
- updating a view of the workspace, based at least in part on said determination, to determine an updated view in which the first object and the second object are separately identified;
- using the first data element to determine an attribute of the first object;
- and
- using the determined attribute of the first object and the updated view to determine a plan to grasp and move the first object.
6 more elements
A computer program product embodied in a non-transitory computer readable medium, comprising computer instructions for:
- receiving from a computer vision system video segmentation data indicating an identified object of interest in a workspace;
- manipulating a robotic arm to move a first sensor to a position such that the identified object of interest is within a read range of the sensor;
- receiving via a communication interface a first sensor data read by the first sensor, the first sensor data comprising a first data element associated with a first object and a second data element associated with a second object;
- receiving via the communication interface a second sensor data read by a second sensor positioned statically in the workspace, the second sensor data comprising the second data element associated with the second object but not the first data element associated with the first object;
- determining based on first sensor data and the second sensor data that the identified object comprises both the first object and the second object;
- updating a view of the workspace, based at least in part on said determination, to determine an updated view in which the first object and the second object are separately identified;
- using the first data element to determine an attribute of the first object;
- and
- using the determined attribute of the first object and the updated view to determine a plan to grasp and move the first object.
6 more elements
Dependencies are inferred from the claim text.