USPTO Grants Patent for Robot Arm Control Device

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 24th, 2026

Summary

The USPTO has granted a new patent (US12583103B2) to the Agency for Defense Development for a robot arm control device and a method for training a hierarchical reinforcement learning model. The patent, granted on March 24, 2026, details a system that uses target images and virtual canvas representations to determine joint angle changes for robot arm operation.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has issued patent US12583103B2 to the Agency for Defense Development. This patent covers a robot arm control device and a method for training a hierarchical reinforcement learning model. The technology described involves using target images and virtual representations to guide robot arm movements, potentially enhancing precision and automation in robotic tasks.

This is a patent grant, which provides exclusive rights to the assignee for their invention. It does not impose new regulatory obligations on other entities. However, companies operating in the robotics and AI sectors, particularly those involved in advanced manufacturing or defense applications, may wish to review the patent details to understand the scope of the granted intellectual property and to inform their own research and development strategies.

Source document (simplified)

← USPTO Patent Grants

Robot arm control device, method for training hierarchical reinforcement learning model for robot arm control, and storage medium storing instructions to perform method training hierarchical reinforcement learning model for robot arm control

Grant US12583103B2 Kind: B2 Mar 24, 2026

Assignee

Agency for Defense Development

Inventors

Ganghun Lee, Byoung Tak Zhang, Minji Kim, Minsu Lee

Abstract

A robot arm control device is proposed. The robot arm control device may include a memory. The device may also include an acquisition unit acquiring an arbitrary target image, a virtual canvas image in which a virtual drawing operation of a robot arm for the target image is reflected, and a virtual nib image of the robot arm whose position is changed as the virtual drawing operation is reflected in the virtual canvas image. The device may further include a processor configured to input the target image to the pre-trained learning model, determine a position change amount of the virtual nib image in the virtual canvas image using the pre-trained learning model, and output a joint angle change amount for driving the robot arm on the basis of the position change amount.

CPC Classifications

B25J 9/163 B25J 9/1664 B25J 9/1679 B25J 9/1671 B25J 9/161 B25J 9/1653 B25J 9/1669 B25J 9/1697 B25J 11/0075 B25J 19/021 G06N 3/008 G06N 3/045 G06N 3/08 G06N 3/092 G05B 19/4205 G05B 19/4202 G05B 2219/36324