Axonnic's forearm sleeve reads muscle signals to track finger movements and estimate force
Y Combinator startup Axonnic has built a forearm sleeve that uses 64 channels of surface electromyography to decode finger movements and estimate fingertip force without covering the hands. The founders argue that muscle activity captures the effort behind skilled tasks, which video alone misses, making the data more useful for training dexterous robots. In a test reported on its benchmarks page, the decoder trained on earlier data showed about 6.9 degrees of average error across 20 joint angles compared with Apple Vision Pro hand tracking, even after the sleeve was removed and reworn. The company says it has collected data from sewing at a European manufacturer, which signed a letter of intent for robotic implementation support, and from electricians, plumbers and demolition workers. The team, which grew out of Oxford research, is also developing its own wired circuit board to reduce interference seen in factory recordings.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in