Back to results
US 2023/0405820 A1Published application
Predicting grasp outcomes
AssigneeNVIDIA Corporation
InventorsIsabella Huang, Yashraj Narang, Tucker Ryer Hermans, Fabio Tozeto Ramos, Ankur Handa, Miles Andrew Macklin, Dieter Fox
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques to generate a predicted outcome of an object resulting from a robotic component applying a force. In at least one embodiment, a predicted outcome of an object resulting from a robotic component applying a force is generated based on, for example, a neural network.
Claims20
A system, comprising:
- at least one processor;
- and
- at least one memory comprising instructions that, in response to execution by the at least one processor, cause the system to at least:
- obtain a mesh of an object, a mesh of a robotic component, and an indication of a force;
- cause a neural network to process the mesh of the object, the mesh of the robotic component, and the indication of the force to generate a representation of the object, the robotic component, and the force;
- and
- cause the neural network to process the representation to generate a predicted outcome of the object resulting from the robotic component applying the force on the object.
4 more elements
A method, comprising:
- obtaining data comprising a set of meshes and an indication of a force, wherein the set of meshes are associated with an object and a robotic component;
- causing a graph neural network to process the data to generate a representation of the data;
- and
- obtaining, from the graph neural network and based on the representation, a predicted outcome of the object resulting from the robotic component applying the force to the object.
1 more element
A non-transitory computer-readable medium comprising instructions that, when performed by at least one processor of a computing device, cause the computing device to at least:
- obtain a mesh of an object, a mesh of a robotic component, and an indication of a force;
- cause a neural network to process the mesh of the object, the mesh of the robotic component, and the indication of the force to generate a graph representation;
- and
- cause the neural network to process the graph representation to generate a predicted outcome of the object resulting from the force applied by the robotic component.
1 more element
Dependencies are inferred from the claim text.