Leadership and Companies
smartARM Uses AI to Remove Manual Grip Switching in Bionic Arms
Toronto startup smartARM is developing a vision-led prosthetic arm that automatically selects grips for everyday objects.
What happened
Toronto-based smartARM is developing a vision-first bionic arm prototype that uses a camera embedded in its palm to identify everyday objects and select suitable grips automatically. Its software uses DINOv2, Meta’s open-source vision model, to recognize objects from a few reference photos, reducing adaptation from weeks to almost instantly. Optional Meta AI Glasses provide an additional first-person view and let users add objects through a phone app. The system is designed to avoid manual switching between grip patterns. Former NFL player and smartARM user Shaquem Griffin said, “scalability and accessibility matter most to me.”
Why it matters
The prototype could make everyday prosthetic use more intuitive by removing a multistep grip-selection process and reducing the learning required to handle different objects. Personalisation also comes through the app and added object recognition.
Source: Meta Newsroom