公开(公告)号：US20230415336A1

公开(公告)日：2023-12-28

申请号：US17849861

申请日：2022-06-27

Applicant: NVIDIA Corporation

Inventor： Siddha Ganju ,  Elad Mentovich ,  James Stephen Fields, JR. ,  Ryan Kelsey Albright ,  Jonathan Tremblay ,  Stanley Thomas Birchfield

IPC: B25J9/16 ,  G05B19/4155

CPC classification number: B25J9/163 ,  G05B19/4155 ,  G05B2219/50391 ,  G05B2219/32226

Abstract: A robot device determines an error associated with equipment included in a data center environment. The robot device may compare the error to candidate errors for which the robot device is already trained to resolve. Based on a result of the comparison, the robot device may perform, in a control environment, candidate maintenance operations in association with resolving the error. The robot device may learn a set of actions associated with successfully resolving the error, based on performing the candidate maintenance operations. The robot device may perform maintenance operations associated with the error. Performing the maintenance operations may include applying the learned set of actions.

公开(公告)号：US20240061388A1

公开(公告)日：2024-02-22

申请号：US17888834

申请日：2022-08-16

Applicant: Nvidia Corporation

Inventor： Siddha Ganju ,  Elad Mentovich ,  James Stephen Fields, JR. ,  Nathan D. Ratliff ,  Ryan Kelsey Albright

IPC: G05B17/02 ,  G05B23/02

CPC classification number: G05B17/02 ,  G05B23/0283

Abstract: A virtual representation of a physical environment can be generated through simulation, which can include one or more virtual agents to represent robots, or at least semi-automated devices, that can operate and perform various tasks in the physical environment. Various component failures, or other potential problems, can be simulated that can be analyzed by one or more deep learning models associated with the virtual agents. These deep learning models can attempt to diagnose the simulated problem, as well as determine one or more potential solutions. The virtual agents can help to gather information for these determinations, as well as to perform tasks for these potential solutions. Once these deep learning models are trained in this simulated environment, these models can be used by one or more robots to perform tasks that may relate to maintenance or operation of a physical environment.