The Benefits of Immersive Demonstrations for Teaching Robots

Astrid Jackson, Brandon D. Northcutt, Gita Sukthankar

Abstract

One of the advantages of teaching robots by demonstration is that it can be more intuitive for users to demonstrate rather than describe the desired robot behavior. However, when the human demonstrates the task through an interface, the training data may inadvertently acquire artifacts unique to the interface, not the desired execution of the task. Being able to use one’s own body usually leads to more natural demonstrations, but those examples can be more difficult to translate to robot control policies.

This paper quantifies the benefits of using a virtual reality system that allows human demonstrators to use their own body to perform complex manipulation tasks. We show hat our system generates superior demonstrations for a deep neural network without introducing a correspondence problem. The effectiveness of this approach is validated by comparing the learned policy to that of a policy learned from data collected via a conventional gaming system, where the user views the environment on a monitor screen, using a Sony Play Station 3 (PS3) DualShock 3 wireless controller as input.

Paper: Published at ACM/IEEE International Conference on Human-Robot Interaction from March 11 - 14, 2019.

Download: Video (720p, mp4, 89MB)

Tasks performed during the user study

Cleanup Task

With conventional gaming system

ps3_cleanup.mp4

In virtual reality

vr_cleanup.mp4

Handover Task

With conventional gaming system

ps3_handover.mp4

In virtual reality

vr_handover.mp4

Blockstacking Task

With conventional gaming system

ps3_blockstacking.mp4

In virtual reality

vr_blockstacking.mp4

Subjective evaluation of the task performance

Participants were asked to give subjective feedback on their task performance. Please find in the following the statistical evaluation of the subjective measures.

user_study_questionnaire.pdf