IP Library Granted Patent US 10,606,898
Granted Patent B2
US 10,606,898 · App. 15/957,651 · Granted Mar 31, 2020

Interpreting human-robot instructions

Inventors: Stefanie Tellex (Cambridge, MA); Dilip Arumugam (Providence, RI); Siddharth Karamcheti (Los Altos Hills, CA); Nakul Gopalan (Providence, RI); Lawson L. S. Wong (SAR, HK)
Assignee: Brown University
G06F16/90332G06F15/76G06F17/2809G06N3/008G06N3/0445G06N3/0454G06N3/08G06N7/005G06N20/00G10L2015/223G10L2015/225
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,606,898
App. No.
15/957,651
Granted
Mar 31, 2020
Kind
B2
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.

Claims (29)

1. A system comprising:

a robot comprising a module to interpret natural language commands, the module comprising 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 comprising:

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;

an Abstract Markov Decision Process to represent a decision-making problem involving a hierarchy of the robot's states and actions, the Abstract Markov Decision Process comprising:

a set of states that define an environment specified in an object-oriented fashion with object classes and attributes;

a set of actions an agent can execute to transition between states or to invoke a lower-level AMDP subtask;

a transition probability distribution over all possible next states given a current state and executed action;

a numerical reward earned for a particular transition;

a discount factor or effective time horizon under consideration; and

a state projection function that maps lower-level states to higher-level (AMDP) states.

2. A system comprising:

a robot comprising a module comprising 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 comprising:

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;

an Abstract Markov Decision Process to represent a decision-making problem involving a hierarchy of the robot's states and actions, the Abstract Markov Decision Process comprising:

a set of states that define an environment specified in an object-oriented fashion with object classes and attributes;

a set of actions an agent can execute to transition between states or to invoke a lower-level AMDP subtask;

a transition probability distribution over all possible next states given a current state and executed action;

a numerical reward earned for a particular transition;

a discount factor or effective time horizon under consideration; and

a state projection function that maps lower-level states to higher-level (AMDP) states.

3. A method comprising:

providing a mobile-manipulator robot; and

providing a module that interprets and grounds natural language commands to a mobile-manipulator robot at multiple levels of abstraction, the module comprising a deep neural network language model that learns how to map the natural language commands to reward functions at an appropriate level of a hierarchical planning framework, the functions comprising an Abstract Markov Decision Process to represent a decision-making problem involving a hierarchy of the robot's states and actions, the Abstract Markov Decision Process comprising:

a set of states that define an environment specified in an object-oriented fashion with object classes and attributes;

a set of actions an agent can execute to transition between states or to invoke a lower-level AMDP subtask;

a transition probability distribution over all possible next states given a current state and executed action;

a numerical reward earned for a particular transition;

a discount factor or effective time horizon under consideration; and

a state projection function that maps lower-level states to higher-level (AMDP) states.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 2, 2020
From: WONG, LAWSON L.S.
To: BROWN UNIVERSITY
Reel/Frame 051976/0358 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 27, 2020
From: ARUMUGAM, DILIP; KARAMCHETI, SIDDHARTH; GOPALAN, NAKUL
To: BROWN UNIVERSITY
Reel/Frame 051634/0321 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 24, 2019
From: TELLEX, STEFANIE
To: BROWN UNIVERSITY
Reel/Frame 050813/0850 →
CONFIRMATORY LICENSE Recorded Apr 25, 2018
From: BROWN UNIVERSITY
To: NATIONAL SCIENCE FOUNDATION
Reel/Frame 046016/0611 →
Continuity (2)
Provisional Application 62487294 · Apr 19, 2017
Related Publication 20180307779A1 · Oct 25, 2018
Cited By (2)
US 12,196,127 US 12,424,110