Search papers, labs, and topics across Lattice.
DexEXO, a novel hand exoskeleton, prioritizes wearability and cross-user adaptability through a pose-tolerant thumb mechanism and slider-based finger interface, accommodating a wide range of hand sizes. By visually matching the deployed robot with a passive hand, DexEXO enables direct policy training from raw RGB observations, eliminating the need for visual post-processing. Diffusion policies trained using this simplified pipeline achieve competitive performance, demonstrating that hardware-level embodiment alignment reduces both human and algorithmic bottlenecks.
A wearable hand exoskeleton that prioritizes comfort and adaptability unlocks scalable robot learning by enabling direct policy training from raw visual data, bypassing complex post-processing.
Scaling dexterous robot learning is constrained by the difficulty of collecting high-quality demonstrations across diverse operators. Existing wearable interfaces often trade comfort and cross-user adaptability for kinematic fidelity, while embodiment mismatch between demonstration and deployment requires visual post-processing before policy training. We present DexEXO, a wearability-first hand exoskeleton that aligns visual appearance, contact geometry, and kinematics at the hardware level. DexEXO features a pose-tolerant thumb mechanism and a slider-based finger interface analytically modeled to support hand lengths from 140~mm to 217~mm, reducing operator-specific fitting and enabling scalable cross-operator data collection. A passive hand visually matches the deployed robot, allowing direct policy training from raw wrist-mounted RGB observations. User studies demonstrate improved comfort and usability compared to prior wearable systems. Using visually aligned observations alone, we train diffusion policies that achieve competitive performance while substantially simplifying the end-to-end pipeline. These results show that prioritizing wearability and hardware-level embodiment alignment reduces both human and algorithmic bottlenecks without sacrificing task performance. Project Page: https://dexexo-research.github.io/
