Physics-Driven Data Generation for Contact-Rich Manipulation via Trajectory Optimization

TL;DR

This paper introduces a physics-based data generation pipeline that combines simulation, human demonstrations, and optimization to create versatile datasets for contact-rich robotic manipulation, enabling cross-embodiment transfer and zero-shot policy deployment.

Contribution

The authors develop a novel, efficient data generation pipeline that refines human demonstrations through optimization, allowing for diverse, physically consistent datasets adaptable across different robot embodiments.

Findings

Policies trained on generated datasets perform well on multiple robot types.

Zero-shot deployment achieves high success rates with minimal human input.

Pipeline enables reuse of legacy datasets across hardware configurations.

Abstract

We present a low-cost data generation pipeline that integrates physics-based simulation, human demonstrations, and model-based planning to efficiently generate large-scale, high-quality datasets for contact-rich robotic manipulation tasks. Starting with a small number of embodiment-flexible human demonstrations collected in a virtual reality simulation environment, the pipeline refines these demonstrations using optimization-based kinematic retargeting and trajectory optimization to adapt them across various robot embodiments and physical parameters. This process yields a diverse, physically consistent dataset that enables cross-embodiment data transfer, and offers the potential to reuse legacy datasets collected under different hardware configurations or physical parameters. We validate the pipeline's effectiveness by training diffusion policies from the generated datasets for challenging contact-rich manipulation tasks across multiple robot embodiments, including a floating Allegro hand and bimanual robot arms. The trained policies are deployed zero-shot on hardware for bimanual iiwa arms, achieving high success rates with minimal human input. Project website: https://lujieyang.github.io/physicsgen/.