Prim-LAfD: A Framework to Learn and Adapt Primitive-Based Skills from Demonstrations for Insertion Tasks

TL;DR

Prim-LAfD is a framework that efficiently learns and adapts primitive-based insertion skills from demonstrations, significantly reducing training time and improving generalization for robotic insertion tasks.

Contribution

It introduces a novel framework using black-box optimization and dense rewards to learn and adapt insertion skills from demonstrations, enhancing data efficiency and generalization.

Findings

Learns insertion skills in less than an hour.

Adapts to new insertion tasks in as little as fifteen minutes.

Effective on multiple peg-hole and connector-socket tasks.

Abstract

Learning generalizable insertion skills in a data-efficient manner has long been a challenge in the robot learning community. While the current state-of-the-art methods with reinforcement learning (RL) show promising performance in acquiring manipulation skills, the algorithms are data-hungry and hard to generalize. To overcome the issues, in this paper we present Prim-LAfD, a simple yet effective framework to learn and adapt primitive-based insertion skills from demonstrations. Prim-LAfD utilizes black-box function optimization to learn and adapt the primitive parameters leveraging prior experiences. Human demonstrations are modeled as dense rewards guiding parameter learning. We validate the effectiveness of the proposed method on eight peg-hole and connector-socket insertion tasks. The experimental results show that our proposed framework takes less than one hour to acquire the insertion skills and as few as fifteen minutes to adapt to an unseen insertion task on a physical robot.