Patent · US Active

Interpreting human-robot instructions

US10606898B2 · kind B2 · utility

7Cited by

1References

3Claims

0Family size

Assignee

Brown University · US

Inventors

Stefanie Tellex · Cambridge, US
Dilip Arumugam · Providence, US
Siddharth Karamcheti · Los Altos Hills, US
Nakul Gopalan · Providence, US
Lawson L. S. Wong · Tuen Mun, CN

Key dates

Filing date	Apr 19, 2018
Grant date	Mar 31, 2020
Priority date	—
Expiry date	Apr 22, 2038

Classification

Technology area (CPC G)Physics
CPC primaryG10L2015/225
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A system includes a robot having a module that includes a function for mapping natural language commands of varying complexities to reward functions at different levels of abstraction within a hierarchical planning framework, the function including using a deep neural network language model that learns how to map the natural language commands to reward functions at an appropriate level of the hierarchical planning framework.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.