Brik H. Meza

Why do people keep collecting data with teleoperation when a bimanual robot setup costs more than $10k? Isn't there a solution that gives you the same data quality without the robot? At @robonet_, we want to build the Internet of Robotics. As part of that mission, we built HandUMI, a hand-worn data collection device for bimanual arms with parallel-jaw grippers. Specs per unit: - 276.5 grams - $110.68 - Encoder-precision gripper aperture - Integrated wrist camera - Tracking with the VR headset of your choice (Pico/Quest) - More than 5 grippers supported (Piper, Trossen, ARX, Soft gripper, Dream gripper) The best part: all the hardware is open source! Thanks @fdotinc for the hardware lab and the space to make this possible. ft. @alvax64 @leoperzz @raulb4s @mbrq_13 @BryanBRstds @Aryan_Mangla_ , and the rest of the @0xnonhuman team.

@nisargxj @NatalieLeviii @obotiq Very cool!

We built a robot to support caregivers, reduce burnout, and enhance care

@DhravyaShah @daleverett is goated. Tomorrow they're going viral.

this right here is a trillion dollar in market value creation

@k7agar @DominiqueCAPaul @raulb4s @mbrq_13 @0xnonhuman @aurel_arnold Every arm with a parallel gripper is planned to be supported. Join the Discord community!

@autobrik @DominiqueCAPaul @raulb4s @mbrq_13 @0xnonhuman @aurel_arnold yam support planned? I can help speed run it

@raulb4s and @mbrq_13 testing teleop with HandUMI at the @0xnonhuman lab in Peru 🇵🇪, 7,250 km away from SF. This is part of the tests to make sure the IK works correctly and that the data collected with HandUMI feels like teleoperated data, but much cheaper. The complete software will be open sourced in the coming days! Data collection, teleop in sim and in real, data postprocessing, etc. Stay tuned! Want to collaborate on HandUMI or build your own? Join our Discord: discord.gg/V47FuUkFA

@DominiqueCAPaul @raulb4s @mbrq_13 @0xnonhuman @aurel_arnold We will check it out. Actually, with this, HandUMI will support data collection and postprocessing for TRLC arms very soon!!

@autobrik @raulb4s @mbrq_13 @0xnonhuman In case you're implementing the IK yourself: @aurel_arnold wrote a very good blog post on how to write good IK and we also open-sourced the code for it. aurelarnold.xyz/blog/vr-teleop… github.com/Dream-Machines…

@yuewang314 Data is key!

It’s all and only about data.

Working hard for the open source community on a 6 yr old laptop with 7 GB RAM, and no GPU. 124 MB available 💀 (4:51 am)

This work focuses on bimanual robotic arms equipped with parallel grippers (Piper, Trossen, ARX, Yam, and so on). We also provide hardware support through ready-to-print gripper designs available in the repository: github.com/BrikHMP18/Hand…, along with future software for post-processing and retargeting data to different robotic arms :)

@autobrik @robonet_ How does it differ from YUBI?

Why do people keep collecting data with teleoperation when a bimanual robot setup costs more than $10k? Isn't there a solution that gives you the same data quality without the robot? At @robonet_, we want to build the Internet of Robotics. As part of that mission, we built HandUMI, a hand-worn data collection device for bimanual arms with parallel-jaw grippers. Specs per unit: - 276.5 grams - $110.68 - Encoder-precision gripper aperture - Integrated wrist camera - Tracking with the VR headset of your choice (Pico/Quest) - More than 5 grippers supported (Piper, Trossen, ARX, Soft gripper, Dream gripper) The best part: all the hardware is open source! Thanks @fdotinc for the hardware lab and the space to make this possible. ft. @alvax64 @leoperzz @raulb4s @mbrq_13 @BryanBRstds @Aryan_Mangla_ , and the rest of the @0xnonhuman team.

Today, for the first time, @dianamunozlar and I are introducing @buivoai 🔥. AI gave every role a faster way to work alone. Business prompts. Product prompts. Engineering prompts. Going in different directions. The problem was never the LLMs. It was the interface. So we built it. The oldest human expression now ships software. The waitlist is open: buivo.ai/?utm_source=tw…

@vineethcyeevani @robonet_ Maybe we can support AprilTags too in the future 👀

@autobrik @robonet_ just use Apriltags. why do you need a full vr controller for this?

@robcoomber @robonet_ Yep, right next to the hardware lab. Swing by anytime, we're here every day!

@autobrik @robonet_ So cool, ur at ft mason?

First of all, @chichengcc 's UMI is an exceptional work. HandUMI differs mainly in: - Ergonomics: you control the gripper using the fingers' natural pinch control instead of a pistol-grip design (More like Generalist style) - Lightweight: 276.5 g + VR controller (125 g), vs. 780 g for original UMI. - Tracking: VR controller tracking instead of SLAM, reducing drift and calibration problems. - Modularity: We are working to support the community's most used robotic arms (with parallel grippers) like Piper, Trossen, ARX, OpenArm, and more.

@autobrik @robonet_ I just learned about UMI today, how does your version differ from the grips I see in this paper here? umi-gripper.github.io

All the hardware is open source, including grippers that fit HandUMI, ready to print: github.com/BrikHMP18/Hand… Instead of using the robot to collect data, USE IT FOR DEPLOYMENT. The world needs more robots outside the labs!

Last year, we were collecting data through teleoperation (left: our old setup). Even though we cut costs with two SO-ARM100 as leaders, it is hard to do dexterous tasks when you cannot feel what you are grasping and you are not close to what you are manipulating. With HandUMI, it is more intuitive because you use your own fingers as the actuator. Right: first video example, plugging a USB-C cable into a keyboard.

Who is struggling with data collection for robotics? Stay tuned tomorrow, and you will save up to 8x on data collection costs. Building at Fuera de temporada II @fdotinc

@DominiqueCAPaul @GeneralistAI Looking for a setup for robot arms with parallel grippers? More soon... stay tuned 👀 github.com/BrikHMP18/Hand…

The new YUBI paper is the closest thing out there to @GeneralistAI's UMI-like grippers. UMI is better than teleoperation but has issues: > Gripper is clunky and heavy ~900g > Uses inaccurate visual SLAM > Pistol trigger closing mechanism feels unnatural YUBI: > Uses Meta Quest for tracking > Is fully open-source > Closing gripper gives user haptic feedback > Weighs 319g - a third of UMI > Open-sourced an 8400h dataset Data collection is ~2x faster than with regular UMI in their usability studies. But what surprised me: After collecting such a huge dataset, they only finetune a few π0.5 policies on minor subsets of the data as POC.

An exciting and long road lies ahead for Physical AI. I’m so excited to be living in this time. I hope that someday we'll see this robot exploring the solar system and beyond.

We're launching the microagi Research Fellowship. Fellows get up to $2M in compute, robotics hardware, our evals, and one of the largest physical AI datasets ever assembled. You build in our lab, with our team, alongside partners like Unitree, Nvidia, and Google Cloud. The hard part of AI left is physical. That's the part we're working on. Come build with us. One more thing: know someone who belongs here? Reply with their name. If they get in, we send you 10.000 USD