RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation

Tianxing Chen, Zanxin Chen, Baijun Chen, Zijian Cai, Yibin Liu, Zixuan Li, Qiwei Liang, Xianliang Lin, Yiheng Ge, Zhenyu Gu, Weiliang Deng, Yubin Guo, Tian Nian, Xuanbing Xie, Qiangyu Chen, KailunSu, Tianling Xu, Guodong Liu, Mengkang Hu, Huan-ang Gao et al. (6 additional authors not shown)

Keywords: Robotics; Embodied AI; Simulation; Sim2Real; Bimanual Manipulation; Synthetic Data Generation

TL;DR: RoboTwin 2.0 is a scalable sim-driven framework that auto-generates diverse bimanual-manipulation data and a unified benchmark, measurably boosting code generation and policy performance.

Abstract: Synthetic data generation via simulation represents a promising approach for enhancing robotic manipulation. However, current synthetic datasets remain insufficient for robust bimanual control due to limited scalability in novel task generation and oversimplified simulations that inadequately capture real-world complexity. We present RoboTwin 2.0, a scalable framework for automated diverse synthetic data generation and unified evaluation for bimanual manipulation. We construct RoboTwin-OD, an object library of 731 instances across 147 categories with semantic and manipulation labels. Building on this, we design a expert data generation pipeline by utilizing multimodal large language models to systhesize task-execution code with simulation-in-the-loop refinement. To improve sim-to-real transfer, RoboTwin 2.0 applies structured domain randomization over five factors (clutter, lighting, background, tabletop height, language instructions). Using this approach, we instantiate 50 bimanual tasks across five robot embodiments. Experimental results demonstrate a 10.9% improvement in code-generation success rates. For downstream learning, vision-language-action models trained with our synthetic data achieve 367% performance improvements in the few-shot setting and 228% improvements in the zero-shot setting, relative to a 10-demo real-only baseline. We further evaluate multiple policies across 50 tasks with two difficulty settings, establishing a comprehensive benchmark to study policy performance. We release the generator, datasets, and code to support scalable research in robust bimanual manipulation.

Primary Area: datasets and benchmarks

Submission Number: 13230