18367827. SYNERGIES BETWEEN PICK AND PLACE: TASK-AWARE GRASP ESTIMATION simplified abstract (Samsung Electronics Co., Ltd.)

SYNERGIES BETWEEN PICK AND PLACE: TASK-AWARE GRASP ESTIMATION

Organization Name

Samsung Electronics Co., Ltd.

Inventor(s)

Nikhil Narsingh Chavan Dafle of Jersey City NJ (US)

Vasileios Vasilopoulos of Woodbridge NJ (US)

Shubham Agrawal of Jersey City NJ (US)

Jinwook Huh of Millburn NJ (US)

Suveer Garg of New York NY (US)

Pedro Piacenza of Jersey City NJ (US)

Isaac Hisano Kasahara of Brooklyn NY (US)

Kazim Selim Engin of Weehawken NJ (US)

Zhanpeng He of New York NY (US)

Shuran Song of New York NY (US)

Ibrahim Volkan Isler of Saint Paul MN (US)

SYNERGIES BETWEEN PICK AND PLACE: TASK-AWARE GRASP ESTIMATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 18367827 titled 'SYNERGIES BETWEEN PICK AND PLACE: TASK-AWARE GRASP ESTIMATION

Simplified Explanation

The patent application describes a system for controlling a robot manipulator to grasp and place objects based on 3D geometry information and affordance information obtained from neural network models.

• The system determines 3D geometry information about a target object and the scene where the object will be placed from images.
• Affordance information is obtained by providing the 3D geometry information to neural network models.
• The robot is commanded to grasp the target object based on the affordance information and to position the manipulator for placing the object in the scene.

Potential Applications

This technology can be applied in industries such as manufacturing, logistics, and healthcare where robots are used for handling objects.

Problems Solved

This technology solves the problem of efficiently and accurately controlling a robot manipulator to grasp and place objects in a given scene.

Benefits

The benefits of this technology include improved automation, increased efficiency, and reduced errors in object manipulation tasks.

Potential Commercial Applications

The potential commercial applications of this technology can be seen in industries that require precise and automated handling of objects, such as warehouses, factories, and hospitals.

Possible Prior Art

One possible prior art for this technology could be robotic systems that use computer vision and neural networks for object recognition and manipulation tasks.

What are the specific neural network models used in this system?

The specific neural network models used in this system are not mentioned in the abstract. Further details about the architecture and training of these models would be needed to understand their specific implementation.

How does the system handle variations in object size and shape?

The abstract does not provide information on how the system handles variations in object size and shape. It would be important to know if the system is capable of adapting to different objects with diverse characteristics for practical applications.

Original Abstract Submitted

Systems, methods, and apparatuses for controlling a robot including a manipulator, including: determining three-dimensional (3D) geometry information about a target object based on an image of the target object; determining 3D geometry information about a scene in which the target object is to be placed based on at least one image of the scene; obtaining affordance information by providing the 3D geometry information about the target object and the 3D geometry information about the scene to at least one neural network model; commanding the robot to grasp the target object using the manipulator according to a grasp orientation corresponding to the affordance information; and commanding the robot to position the manipulator according to a placement direction corresponding to the affordance information in order to place the target object at a location in the scene.