Tag
Axonnic (YC F26) launches a 64-channel EMG forearm sleeve that decodes finger joint angles and fingertip force without touching the hands, aimed at recording skilled manual labor to generate training data for dexterous robots, reporting 6.9° error across 20 joints without recalibration.
Axonnic (YC F26) is launching an EMG sleeve that captures finger movement and force to train robots on real human skill, backed by Y Combinator.
LiteEMG-FM is an efficient hybrid CNN-Transformer foundation model pretrained on 16 EMG datasets for robust, zero-calibration gesture recognition. The paper demonstrates real-time deployability on resource-constrained edge devices via a hierarchical wake-up gate and evaluates latency, power, and memory trade-offs across offloading paradigms.
Preload (YC F26) announced synchronized egocentric video, EMG, and tactile data collection for robot manipulation, VLAs, and world models, offering end-to-end robotic data collection at scale.
This paper presents a learning-based framework that models individual scorer behavior via confusion matrices to derive more robust sleep stage labels from multiple experts, improving accuracy on DOD-H and DOD-O datasets.