Augmentation for Learning From Demonstration with Environmental Constraints

TL;DR

This paper presents a Learning from Demonstration approach that uses autonomous augmentation of a single human demonstration to enable contact-rich manipulation of articulated mechanisms, achieving generalization and robustness in changing environments.

Contribution

It introduces an autonomous augmentation method for LfD that improves generalization and robustness from a single demonstration in contact-rich tasks.

Findings

Successfully generalizes to different mechanisms of the same type

Robustly handles environmental variations in real-world experiments

Achieves reliable task completion in complex, multi-DOF mechanisms

Abstract

We introduce a Learning from Demonstration (LfD) approach for contact-rich manipulation tasks with articulated mechanisms. The extracted policy from a single human demonstration generalizes to different mechanisms of the same type and is robust against environmental variations. The key to achieving such generalization and robustness from a single human demonstration is to autonomously augment the initial demonstration to gather additional information through purposefully interacting with the environment. Our real-world experiments on complex mechanisms with multi-DOF demonstrate that our approach can reliably accomplish the task in a changing environment. Videos are available at the: https://sites.google.com/view/rbosalfdec/home