(YIP) ROBOT LEARNING FROM DEMONSTRATION WITH AUXILIARY CONTEXTUAL DATA
Abstract
Learning from demonstration has emerged as a powerful way to quickly and naturally program robots to perform a wide variety of tasks. Unfortunately, demonstrations are inherently ambiguous and incomplete, making correct generalization to unseen situations difficult without a large number of demonstrations in varying conditions. By contrast, humans are often able to learn complex tasks from a single demonstration by leveraging context learned over a lifetime, such as knowledge about how objects work, episodic memories of similar situations, and an intuitive understanding of the intentions of the demonstrator.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2021
- Source ID
- FA95502010077
Entities
People
- Scott Niekum
Organizations
- Air Force Office of Scientific Research
- United States Air Force
- University of Texas at Austin