Published on:
12 March 2024
Primary Category:
Robotics
Paper Authors:
Shivin Dass,
Wensi Ai,
Yuqian Jiang,
Samik Singh,
Jiaheng Hu,
Ruohan Zhang,
Peter Stone,
Ben Abbatematteo,
Roberto Martin-Martin
TeleMoMa enables teleoperating mobile robots with various human interfaces and their combinations
It facilitates whole-body control using only vision cameras, lowering barrier to entry
TeleMoMa controls different robots like Tiago, HSR and Fetch in simulation and real world
It collects high-quality demos for learning mobile manipulation policies that require coordinated motions
Teleoperation system for mobile robot control
This paper introduces TeleMoMa, a modular and versatile system for teleoperating mobile manipulators. It allows combining multiple interfaces like vision cameras, VR controllers, joysticks, etc. to control different parts of a robot. TeleMoMa lowers barriers to collecting mobile manipulation demonstrations by enabling whole-body control with just vision cameras. The authors demonstrate TeleMoMa by controlling various robots in simulation and the real world. They train policies on TeleMoMa demonstrations that successfully perform complex mobile tasks involving synchronized arm and base motions.
Teleoperation with haptic feedback using low-cost tactile sensors
Teleoperation system for robot control
Learning dexterity from human demonstrations
Robotic system learns new manipulation skills from user demonstrations
Human emotion detection from body motion and context
Learning coordinated mobile manipulation
No comments yet, be the first to start the conversation...
Sign up to comment on this paper