D-CODA: Diffusion for Coordinated Dual-Arm Data Augmentation
This addresses the problem of costly data collection for bimanual manipulation in robotics, offering a scalable augmentation solution, though it is incremental as it extends visual augmentation from single-arm to dual-arm settings.
The paper tackles the challenge of learning bimanual manipulation by proposing D-CODA, a diffusion-based method for offline data augmentation that generates viewpoint-consistent wrist-camera images and joint-space action labels, and it outperforms baselines in 5 simulated and 3 real-world tasks across 2250 simulation and 300 real-world trials.
Learning bimanual manipulation is challenging due to its high dimensionality and tight coordination required between two arms. Eye-in-hand imitation learning, which uses wrist-mounted cameras, simplifies perception by focusing on task-relevant views. However, collecting diverse demonstrations remains costly, motivating the need for scalable data augmentation. While prior work has explored visual augmentation in single-arm settings, extending these approaches to bimanual manipulation requires generating viewpoint-consistent observations across both arms and producing corresponding action labels that are both valid and feasible. In this work, we propose Diffusion for COordinated Dual-arm Data Augmentation (D-CODA), a method for offline data augmentation tailored to eye-in-hand bimanual imitation learning that trains a diffusion model to synthesize novel, viewpoint-consistent wrist-camera images for both arms while simultaneously generating joint-space action labels. It employs constrained optimization to ensure that augmented states involving gripper-to-object contacts adhere to constraints suitable for bimanual coordination. We evaluate D-CODA on 5 simulated and 3 real-world tasks. Our results across 2250 simulation trials and 300 real-world trials demonstrate that it outperforms baselines and ablations, showing its potential for scalable data augmentation in eye-in-hand bimanual manipulation. Our project website is at: https://dcodaaug.github.io/D-CODA/.