Back to results
US 12,591,229 B2Granted
Performance recreation system
AssigneeAivot, LLC
InventorShashwat Srivastav
Life of this patent
Estimating the term…
Abstract
The present disclosure generally relates to performance recreation, and in particular, the recreation of observed human performance using reinforcement learning. In this regard, a first object is identified from a plurality of objects. The manipulation of the first object is tracked from a first position to a second position. A characterization of the manipulation is generated. A policy that controls a mechanical gripper to recreate the manipulation is generated based on an iteratively increasing cumulative award. The mechanical gripper iteratively recreates the manipulation to increase a cumulative award with each recreation.
Claims16
A robotic system comprising:
- an arm assembly;
- a photon detector
- a processor device;
- and
- a non-transitory computer-readable storage medium, coupled with the processor device, having instructions stored thereon, which, when executed by the processor device, cause the processor device to perform actions comprising:
- employing the arm assembly to initiate moving an object away from an initial position and towards a final position within an environment of the robotic system based on a policy learned through at least one previous iteration of the robotic system moving the object or another object between positions within the environment;
- employing the photon detector to acquire image data that depicts at least a current position of the arm assembly and the final position;
- determining, based on the image data, a difference between the current position of the arm assembly and the final position;
- calculating a movement of the arm assembly that traverses the determined difference between the current position of the arm assembly and the final position;
- performing the calculated movement that traverses the determined difference between the current position of the arm assembly and the final position to position the object at the final position;
- and
- updating, based on the calculated movement, the policy learned through the at least one previous iteration of the robotic system moving the object or another object between the positions within the environment by iteratively increasing a cumulative award;
- wherein the cumulative award is based on a plurality of weights associated with each of the initial position, one or more positions of the arm assembly that are intermediate the initial position and the final position, and the final position, and wherein a weight of the plurality of weights that is associated with the final position is larger than the plurality of weights that are associated with the one or more positions of the arm assembly that are intermediate the initial position and the final position.
10 more elements
A method comprising:
- employing an arm assembly of a robot to initiate moving an object away from an initial position and towards a final position within an environment of the robot based on a policy learned through at least one previous iteration of the robot moving the object or another object between positions within the environment;
- employing a photon detector of the robot to acquire image data that depicts at least a current position of the arm assembly and the final position;
- determining, based on the image data, a difference between the current position of the arm assembly and the final position;
- calculating a movement of the arm assembly that traverses the determined difference between the current position of the arm assembly and the final position;
- performing the calculated movement that traverses the determined difference between the current position of the arm assembly and the final position to position the object at the final position;
- and
- updating, based on the calculated movement, the policy learned through the at least one previous iteration of the robot moving the object or another object between the positions within the environment by iteratively increasing a cumulative award;
- wherein the cumulative award is based on a plurality of weights associated with each of the initial position, one or more positions of the arm assembly that are intermediate the initial position and the final position, and the final position, and wherein a weight of the plurality of weights that is associated with the final position is larger than the plurality of weights that are associated with the one or more positions of the arm assembly that are intermediate the initial position and the final position.
5 more elements
One or more non-transitory computer-readable storage media, coupled with a processor device, having instructions stored thereon, which, when executed by the processor device, causes the processor device to perform operations comprising:
- employing an arm assembly of a robot to initiate moving an object away from an initial position and towards a final position within an environment of the robot based on a policy learned through at least one previous iteration of the robot moving the object or another object between positions within the environment;
- employing a photon detector of the robot to acquire image data that depicts at least a current position of the arm assembly and the final position;
- determining, based on the image data, a difference between the current position of the arm assembly and the final position;
- calculating a movement of the arm assembly that traverses the determined difference between the current position of the arm assembly and the final position;
- performing the calculated movement that traverses the determined difference between the current position of the arm assembly and the final position to position the object at the final position;
- and
- updating, based on the calculated movement, the policy learned through the at least one previous iteration of the robot moving the object or another object between positions within the environment by iteratively increasing a cumulative award;
- wherein the cumulative award is based on a plurality of weights associated with each of the initial position, the one or more positions of the arm assembly that are intermediate the initial position and the final position, and the final position, and wherein a weight of the plurality of weights that is associated with the final position is larger than the plurality of weights that are associated with the one or more positions of the arm assembly that are intermediate the initial position and the final position.
5 more elements
Dependencies are inferred from the claim text.