Robots do precise blind assembly using only joint encoders — no cameras, no tactile sensors
KyleMorgenstein · x · 2026-10-20
Researchers show robots already carry contact sensors: their own joints. When touching objects, joints drift slightly off command and encoders pick up the signal.
- Legged robots already use this signal to walk blind; the team shows it also suffices for blind precision assembly
- Policies insert pegs, mesh gears, and thread nuts using only joint positions — no camera, no tactile sensors, no object poses
- The peg policy transfers zero-shot to the real world
Implication: proprioceptive feel, not vision, may be the key to the last-millimeter manipulation problem in general-purpose robotics.
Related event: Robots Achieve Blind Assembly Using Only Joint Encoder Signals(2 posts)→
More from Embodied
- Berkeley EMBER humanoid seminar series: CMU's Guanya Shi on generalist humanoids — GuanyaShi · 2026-10-20
- Apple reportedly set to unveil touchscreen, eSIM MacBook on Oct 28 — xiaohu · 2026-10-20
- Same interface drives a real robot: buzz-wire demo and EmbodiedSafetyExam — ManlingLi_ · 2026-10-20
- RoboEngineer demo: AutoResearch on sim2sim — ManlingLi_ · 2026-10-20
- RoboDynamics demo: real-time tasks where the world won't pause — ManlingLi_ · 2026-10-20
- RoboARC demo: reasoning under spatial constraints — ManlingLi_ · 2026-10-20