Bootstrapping Motor Skill Learning with Motion Planning

TL;DR

This paper introduces a method using kinematic motion planning to autonomously bootstrap motor skill learning in robots, reducing reliance on human demonstrations and improving learning efficiency for complex object manipulation and dynamic tasks.

Contribution

The paper presents a novel approach that leverages motion planning to autonomously initialize motor skills, enabling efficient learning without human input across various complex tasks.

Findings

Motion planning effectively bootstraps motor skills in manipulation tasks.

The method outperforms random policy initialization and matches human demonstration in effectiveness.

It enables autonomous learning of dynamic tasks like hitting a ball off a tee.

Abstract

Learning a robot motor skill from scratch is impractically slow; so much so that in practice, learning must be bootstrapped using a good skill policy obtained from human demonstration. However, relying on human demonstration necessarily degrades the autonomy of robots that must learn a wide variety of skills over their operational lifetimes. We propose using kinematic motion planning as a completely autonomous, sample efficient way to bootstrap motor skill learning for object manipulation. We demonstrate the use of motion planners to bootstrap motor skills in two complex object manipulation scenarios with different policy representations: opening a drawer with a dynamic movement primitive representation, and closing a microwave door with a deep neural network policy. We also show how our method can bootstrap a motor skill for the challenging dynamic task of learning to hit a ball off a tee, where a kinematic plan based on treating the scene as static is insufficient to solve the task, but sufficient to bootstrap a more dynamic policy. In all three cases, our method is competitive with human-demonstrated initialization, and significantly outperforms starting with a random policy. This approach enables robots to to efficiently and autonomously learn motor policies for dynamic tasks without human demonstration.