PRIME: Scaffolding Manipulation Tasks with Behavior Primitives for Data-Efficient Imitation Learning

TL;DR

PRIME introduces a primitive-based framework that decomposes complex manipulation tasks into sequences of behavior primitives, significantly enhancing data efficiency and success rates in imitation learning for robots.

Contribution

The paper presents PRIME, a novel primitive-based imitation learning framework that improves data efficiency and performance in long-horizon manipulation tasks.

Findings

Achieves 10-34% higher success rates in simulation.

Achieves 20-48% higher success rates on physical hardware.

Demonstrates improved performance over state-of-the-art baselines.

Abstract

Imitation learning has shown great potential for enabling robots to acquire complex manipulation behaviors. However, these algorithms suffer from high sample complexity in long-horizon tasks, where compounding errors accumulate over the task horizons. We present PRIME (PRimitive-based IMitation with data Efficiency), a behavior primitive-based framework designed for improving the data efficiency of imitation learning. PRIME scaffolds robot tasks by decomposing task demonstrations into primitive sequences, followed by learning a high-level control policy to sequence primitives through imitation learning. Our experiments demonstrate that PRIME achieves a significant performance improvement in multi-stage manipulation tasks, with 10-34% higher success rates in simulation over state-of-the-art baselines and 20-48% on physical hardware.