GELLO
A low-cost, open-source teleoperation system for robot arms from UC Berkeley — a scaled, kinematically matched replica arm for collecting demonstrations.
#teleoperation#imitation learning#UC Berkeley#data collection#robot arm#open source
Market classResearch
Classification
Embodied data-collection rig
Which group GELLO belongs to and how it is built
Control
Hardware domain
Research
Market class
Teleoperation Interfaces
Subcategory
Subcategory covering hardware teleoperation interfaces: motion capture gloves, motion controllers and wearables that map operator motion to robot actions and generate data for manipulation learning.
Embodied data-collection rig
Component type
A component type: a complete hardware rig for embodied data collection — combining cameras (first-person view), tactile and audio sensors and UMI-standard grippers, with frame-level synchronization. Enables fast, low-cost collection of human demonstrations without a robot.
1 mention of GELLO
Updated: 5 September 2026