Local Policies Enable Zero-shot Long-horizon Manipulation

TL;DR

ManipGen introduces local policies for sim2real robotic manipulation, enabling zero-shot long-horizon tasks with high success rates and outperforming existing methods across numerous real-world benchmarks.

Contribution

The paper proposes local policies for sim2real transfer, demonstrating state-of-the-art zero-shot performance and robustness in long-horizon manipulation tasks.

Findings

Achieves 97% success in simulation benchmarks.

Solves unseen long-horizon tasks with up to 8 stages.

Outperforms SOTA methods by significant margins across 50 real-world tasks.

Abstract

Sim2real for robotic manipulation is difficult due to the challenges of simulating complex contacts and generating realistic task distributions. To tackle the latter problem, we introduce ManipGen, which leverages a new class of policies for sim2real transfer: local policies. Locality enables a variety of appealing properties including invariances to absolute robot and object pose, skill ordering, and global scene configuration. We combine these policies with foundation models for vision, language and motion planning and demonstrate SOTA zero-shot performance of our method to Robosuite benchmark tasks in simulation (97%). We transfer our local policies from simulation to reality and observe they can solve unseen long-horizon manipulation tasks with up to 8 stages with significant pose, object and scene configuration variation. ManipGen outperforms SOTA approaches such as SayCan, OpenVLA, LLMTrajGen and VoxPoser across 50 real-world manipulation tasks by 36%, 76%, 62% and 60% respectively. Video results at https://mihdalal.github.io/manipgen/