公开(公告)号：US20240131706A1

公开(公告)日：2024-04-25

申请号：US18200347

申请日：2023-05-22

Applicant: NVIDIA Corporation

Inventor： Balakumar Sundaralingam ,  Siva Kumar Sastry Hari ,  Adam Harper Fishman ,  Caelan Reed Garrett ,  Alexander James Millane ,  Elena Oleynikova ,  Ankur Handa ,  Fabio Tozeto Ramos ,  Nathan Donald Ratliff ,  Karl Van Wyk ,  Dieter Fox

IPC: B25J9/16

CPC classification number: B25J9/1664

Abstract: Apparatuses, systems, and techniques to perform collision-free motion generation (e.g., to operate a real-world or virtual robot). In at least one embodiment, at least a portion of the collision-free motion generation is performed in parallel.

公开(公告)号：US20210334630A1

公开(公告)日：2021-10-28

申请号：US16860486

申请日：2020-04-28

Applicant: NVIDIA Corporation

Inventor： Alexander Conrad Lambert ,  Adam Harper Fishman ,  Dieter Fox ,  Byron Boots ,  Fabio Tozeto Ramos

IPC: G06N3/04 ,  G06N5/04 ,  G06N3/063 ,  G06F17/18 ,  G06K9/62 ,  G05D1/00

Abstract: Apparatuses, systems, and techniques to infer a sequence of actions to perform using one or more neural networks trained, at least in part, by optimizing a probability distribution function using a cost function, wherein the probability distribution represents different sequences of actions that can be performed. In at least one embodiment, a model predictive control problem is formulated as a Bayesian inference task to infer a set of solutions.

公开(公告)号：US20240037367A1

公开(公告)日：2024-02-01

申请号：US18133986

申请日：2023-04-12

Applicant: NVIDIA Corporation

Inventor： Alexander Conrad Lambert ,  Adam Harper Fishman ,  Dieter Fox ,  Byron Boots ,  Fabio Tozeto Ramos

IPC: G06N3/006 ,  G05D1/00 ,  G06F17/18 ,  G06N3/063 ,  G06N5/04 ,  G06F18/214 ,  G06N3/047

CPC classification number: G06N3/006 ,  G05D1/0088 ,  G06F17/18 ,  G06N3/063 ,  G06N5/04 ,  G06F18/214 ,  G06N3/047

Abstract: Apparatuses, systems, and techniques to infer a sequence of actions to perform using one or more neural networks trained, at least in part, by optimizing a probability distribution function using a cost function, wherein the probability distribution represents different sequences of actions that can be performed. In at least one embodiment, a model predictive control problem is formulated as a Bayesian inference task to infer a set of solutions.

公开(公告)号：US11645492B2

公开(公告)日：2023-05-09

申请号：US16860486

申请日：2020-04-28

Applicant: NVIDIA Corporation

Inventor： Alexander Conrad Lambert ,  Adam Harper Fishman ,  Dieter Fox ,  Byron Boots ,  Fabio Tozeto Ramos

IPC: G06F17/18 ,  G06N3/04 ,  G05D1/00 ,  G06K9/62 ,  G06N3/063 ,  G06N5/04

CPC classification number: G06N3/0472 ,  G05D1/0088 ,  G06F17/18 ,  G06K9/6256 ,  G06N3/063 ,  G06N5/04

Abstract: Apparatuses, systems, and techniques to infer a sequence of actions to perform using one or more neural networks trained, at least in part, by optimizing a probability distribution function using a cost function, wherein the probability distribution represents different sequences of actions that can be performed. In at least one embodiment, a model predictive control problem is formulated as a Bayesian inference task to infer a set of solutions.