Scalable Multi-Task Data Generation via Reinforcement Learning for Language-Conditioned Bimanual Dexterous Manipulation
For researchers in robotic manipulation, this provides a scalable method to generate training data without human teleoperation, addressing a key bottleneck in learning generalist policies.
The paper addresses the lack of large-scale, high-quality datasets for bimanual dexterous manipulation by proposing a reinforcement learning-based data generation pipeline that produces diverse, language-annotated datasets, enabling multi-task policy training with improved generalization across three tasks.
A key bottleneck in training generalist policies for bimanual dexterous manipulation is the lack of large-scale, high-quality datasets. Synthetic data generation in simulation provides a scalable alternative to human video demonstrations by overcoming challenges such as morphology mismatch, missing physical interactions, and the generation of robot actions. However, existing approaches based on human teleoperation offer limited task diversity, as object-centric trajectory matching often neglects the feasibility of robot execution. Reinforcement learning (RL) enables broader scalability but is often constrained by handcrafted, task-specific rewards. In this work, we propose a systematic RL-based data generation pipeline that integrates generalizable reward design, effective domain randomization, and language-conditioned task annotations. This pipeline synthesizes diverse, high-quality datasets for dexterous bimanual manipulation and enables training of language-conditioned multi-task policies. Our experiments show that the generated data significantly improves generalization across three representative manipulation tasks.