Robotics researcher argues proprioception, not vision, is the real bottleneck for grasping
arankomatsuzaki · x · 2026-09-28
Aran Komatsuzaki makes a counterintuitive observation: blind people can still do the majority of tasks where robots fail, suggesting the bottleneck for embodied AI may be proprioception (plus tactile sensing) rather than vision.
He points to work using a "see to reach, feel to grasp" pipeline: camera/VLM/tracker places the palm approximately, then a blind hand reflex confirms the grasp ("yes, I've got it") while the arm continues the task. This achieved robust blind grasping across diverse objects without ever observing contact geometry.
Related event: Robots may need proprioception, not better vision, to grasp(2 posts)→
More from Embodied
- Quadruped Gait Atlas 3D: Muscle-Driven Skeletons Solve Lowest-Cost Gaits at Every Speed — burny_tech · 2026-09-28
- AnyMo, a Setup-Agnostic IMU Human Motion Model, Accepted at NeurIPS 2026 — flosalim · 2026-09-28
- Leslie Kaelbling on Why Astra Excels at Robotics—and Why It Isn't Solved Yet — shaohua0116 · 2026-09-28
- Meta's official XR FOV Simulator used for accurate Quest 3 vs VR glasses FOV comparison — ZeroStateReflex · 2026-09-28
- Dev builds own Playdate visual simulator with Figma-to-dither workflow — eschadiol · 2026-09-28
- InternW0-Δ open-sources code, weights and 20K+ hours of robot data — arankomatsuzaki · 2026-09-28