Patents by Inventor Ajay Uday Mandlekar

Ajay Uday Mandlekar has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Controlling position of robot by determining goal proposals by using neural networks
    Patent number: 11958529
    Abstract: A framework for offline learning from a set of diverse and suboptimal demonstrations operates by selectively imitating local sequences from the dataset. At least one embodiment recovers performant policies from large manipulation datasets by decomposing the problem into a goal-conditioned imitation and a high-level goal selection mechanism.
    Type: Grant
    Filed: August 20, 2020
    Date of Patent: April 16, 2024
    Assignee: NVIDIA CORPORATION
    Inventors: Ajay Uday Mandlekar, Fabio Tozeto Ramos, Byron Boots, Animesh Garg, Dieter Fox
  • MACHINE LEARNING USING MODULAR SOLUTION DATASETS
    Publication number: 20220055689
    Abstract: A framework for offline learning from a set of diverse and suboptimal demonstrations operates by selectively imitating local sequences from the dataset. At least one embodiment recovers performant policies from large manipulation datasets by decomposing the problem into a goal-conditioned imitation and a high-level goal selection mechanism.
    Type: Application
    Filed: August 20, 2020
    Publication date: February 24, 2022
    Inventors: Ajay Uday Mandlekar, Fabio Tozeto Ramos, Byron Boots, Animesh Garg, Dieter Fox