Access to *diverse* training data is a major bottleneck in robot learning. We're releasing DROID, a large-scale in-the-wild manipulation dataset. 76k trajectories, 500+ scenes, multi-view stereo, language annotations etc
Check it out & download today!
💻: https://t.co/JsbBZIxzZA
So proud of my team’s work! Check out this cool video summarizing what we’ve been up to:
https://t.co/9QXLUh1MVJ
and the blog post for more details:
https://t.co/5gzU2yLZNi
How can we collect deployable, in-the-wild, complex robot manipulation data without the need for in-the-wild robots?
Introducing Universal Manipulation Interface (UMI)!
Check out our website to see what you can do with our open-sourced system! https://t.co/ziWBRTcJOJ
Can we collect robot data without any robots?
Introducing Universal Manipulation Interface (UMI)
An open-source $400 system from @Stanford designed to democratize robot data collection
0 teleop -> autonomously wash dishes (precise), toss (dynamic), and fold clothes (bimanual)
Too lazy to pack your laundry or your shopping bag? We introduce a real-world robotics system that takes a step toward this goal!
Website: https://t.co/8RU6dBgqvT
arXiv: https://t.co/tC5l5c1g4m
Video: https://t.co/7Rl5ThLODG
@allieonthebeach @OpenRoboticsOrg Howdy! Sounds like may be current working directory issue if you're using relative path? Agreed w/ @OpenRoboticsOrg that issue with exact repro steps would help community help you.
@soumithchintala@awscloud For Drake (https://t.co/ffRI36OWI4), we have switched from Colab to @DeepnoteHQ. Though I don't think there is a free GPU tier (yet ;).
Since 2021, TRI's University Collaborative Research Program has launched 34 projects, including collaborations with @Columbia, @Stanford, @CarnegieMellon, @UCBerkeley, and @UMich.
Learn how these projects can change your future for the better:
https://t.co/GfAUdE0fsR
@jperras@dcousineau On the thread of moving in a college town: Always fun to be an observer (not participant) for Allston Christmas: https://t.co/T0hY1GKPzp