Quantifying Hypothesis Space Misspecification in Learning from Human-Robot Demonstrations and Physical Corrections

Andreea Bobu, Andrea Bajcsy, Jaime F. Fisac, Sampada Deglurkar, Anca D. Dragan

Research output: Contribution to journalArticlepeer-review

27 Scopus citations

Abstract

The human input has enabled autonomous systems to improve their capabilities and achieve complex behaviors that are otherwise challenging to generate automatically. Recent work focuses on how robots can use such inputs-such as, demonstrations or corrections-to learn intended objectives. These techniques assume that the human's desired objective already exists within the robot's hypothesis space. In reality, this assumption is often inaccurate: there will always be situations where the person might care about aspects of the task that the robot does not know about. Without this knowledge, the robot cannot infer the correct objective. Hence, when the robot's hypothesis space is misspecified, even methods that keep track of uncertainty over the objective fail because they reason about which hypothesis might be correct, and not whether any of the hypotheses are correct. In this article, we posit that the robot should reason explicitly about how well it can explain human inputs given its hypothesis space and use that situational confidence to inform how it should incorporate the human input. We demonstrate our method on a 7 degrees-of-freedom robot manipulator in learning from two important types of human inputs: demonstrations of motion planning tasks and physical corrections during the robot's task execution.

Original languageEnglish (US)
Article number9007490
Pages (from-to)835-854
Number of pages20
JournalIEEE Transactions on Robotics
Volume36
Issue number3
DOIs
StatePublished - Jun 2020
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Keywords

  • Bayesian inference
  • inverse reinforcement learning (IRL)
  • learning from demonstration
  • physical human-robot interaction

Fingerprint

Dive into the research topics of 'Quantifying Hypothesis Space Misspecification in Learning from Human-Robot Demonstrations and Physical Corrections'. Together they form a unique fingerprint.

Cite this