ETH student documents building and calibrating a UMI gripper from scratch
SongShuran · x · 2026-10-07
ETH Zurich undergrad Gino Kreis publishes a detailed blog on building and calibrating his Dream-UMI gripper — a handheld UMI device for collecting robot training data on the TRLC-DK1 arm, meant to complement hard-to-scale teleoperation.
Key points:
- Milestone 1 reached: capture-ready device with camera and encoder streaming over USB, Quest over Wi-Fi, calibration done
- Next milestones: bimanual mobile collection system with robust post-processing, and integrating UMI data into training (pre-train on UMI, fine-tune on tele-op data)
- Covers details papers skip: ArUco tags vs encoder trade-offs, VR controller calibration, SLAM-based pose tracking
More from Embodied
- 7 minutes per motor, ~$5 labor cost: why robotics automation is the only path for Western manufacturing — avlok · 2026-10-07
- Reward-DAgger: generalist reward models enable task-agnostic runtime monitoring for robots — ebiyik_ · 2026-10-07
- NVIDIA's VeriFine Co-Evolves Policy and Judge to Scale Self-Improvement in Embodied Reasoning — nvidia · 2026-10-07
- EmbodiedSmith: Recursive Self-Improvement Flywheel Scales Embodied Training Data in Simulation — Yikai Qin · 2026-10-07
- Kangwook Lee on how games can help build AI for the physical world — Kangwook_Lee · 2026-10-07
- Tencent Hunyuan releases WorldPlay2, a consistent interactive world model steerable by prompts — Scobleizer · 2026-10-07