Researchers at Brown University have developed a system designed to help robots better follow spoken instructions.
The research was led by Dilip Arumugam and Siddharth Karamcheti in the lab of computer science professor Stefanie Tellex.
The system was developed using Mechanical Turk, Amazon's crowdsourcing marketplace, and a virtual task domain called Cleanup World. Volunteers were asked what instructions they would give the robot to get it to perform a task they had just viewed. The volunteers were given guidance as to the level of specificity their directions should have.
The researchers used the volunteers' spoken instructions to train their system to understand what kinds of words are used in each level of abstraction. From there, the system learned to infer not only a desired action, but also the abstraction level of the command, which triggered its hierarchical planning algorithm to solve a task from the appropriate level.
From News from Brown
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found