Juanbe

Profe de fp. Sí a la robótica agrícola y gastronómica. cultivo aguacates!

[Juanbe @juanbebg]

@damad5jaque Korea 2002

[Juanbe @juanbebg]

@anitakedavra Puedes poner en contras de la FP, que te puede tocar dar FP Básica. Lo peor de la ESO en un concentrado.

[Nav Toor @heynavtoor]

🚨 Someone just open sourced an operating system for robots. Any robot. One framework.

It's called DimOS. An open source OS that lets you program humanoids, drones, robot dogs, and robotic arms the same way you'd write a Python script.

No PhD in robotics. No custom firmware. Just code.

Here's what it controls:

→ Quadrupeds: Unitree Go2, Unitree B1, AGIBOT, Dobot
→ Drones: DJI Mavic, Holybro x500
→ Humanoids: Unitree G1, Booster K1, AGIBOT X2
→ Robotic Arms: xARM, AgileX Piper, OpenARMs

All from one single framework.

Here's the wildest part:

Connect Cursor or Claude Code to your robot via MCP. Type "move forward 1 meter." The robot moves. You're vibe-coding a real machine.

That's it. You now have a robot operating system.

Robotics labs spend millions building this kind of software stack.

100% Open Source.

[Erick @ErickSky]

🚨@norma_core_dev acaba de entrar en PUBLIC ALPHA y está destinado a cambiar la robótica para siempre

Esta es una plataforma open-source unificada para Physical AI. En vez de pelearte con 10 librerías diferentes, aquí tienes TODO en un solo stack de alto rendimiento:

- Control de motores.
- Sensores.
- Calibración.
- Teleoperación.
- Imitation learning.
- VLA models.
- Todo junto y optimizado.

Lo más brutal del proyecto:
[ElRobot]

- Brazo robótico de 7+1 grados de libertad (7 articulaciones + gripper).
- 100% imprimible en 3D (archivos listos para Bambu Lab incluidos).
- Cuesta solo ~$220.
- Viene con CAD, URDF para simulación, guía de montaje completa y todo listo para Physical AI.

Es literalmente el “Raspberry Pi de los brazos robóticos”

La novedad:

- Auto-CalibrationAcaban de lanzar calibración automática para CUALQUIER plataforma robótica.
- Tienes dos formas de usarla: ✨ Interfaz web súper bonita 💻 API Pythonic (la que ellos llaman “the true way”)
- Y el código es tan limpio que da gusto copiarlo.

¿Qué está pasando exactamente en el video?

- Hay dos ElRobot reales sobre una mesa (uno gris con gripper grande y uno blanco con cámara en la muñeca).
- En la laptop se está ejecutando el script de auto-calibración (el código que ves en la imagen).

Cada robot se calibra solo, sin que nadie lo toque:

- Mueve una articulación por vez.
- Busca el límite máximo y mínimo.
- Mide torque en tiempo real.
- Va a posiciones float con control preciso.
- Todo en paralelo y súper suave.

En el terminal ves los logs en vivo:

- “Finding max position”, “Torque after: 1.0 Nm”, “Motor calibrated”, etc.

- Y mientras, los robots giran hombros, codos, muñecas y grippers como por arte de magia.

- Al final vuelven a posición neutral y el código dice “Calibration complete”.

¡Es la prueba perfecta de que la librería funciona de verdad en hardware físico!

¿Para quién es esto?

- Makers y hackers que quieren armar su propio brazo barato.

- Investigadores de robótica y Physical AI.

- Cualquiera que quiera experimentar con imitation learning sin gastarse miles de dólares.

Backend en Go (ultra rápido) + API Python preciosa.
Comunidad muy activa en Discord.
¿QUÉ MÁS QUIERESSSSSS?

REPOOO👇

[Gain of Fauci @DschlopesIsBack]

I’m going to continue posting this video every week for the rest of my life 😂🤣

[Asimov @asimovinc]

You can build your own humanoid at home. Asimov – Here be Dragons is now available for presale. $499 deposit, $15,000 target price. https://t.co/DMIO7Qaj76

[Tom Dörr @tom_doerr]

3D-printable rack system for hardware

https://t.co/wFs4PEseKu

[Ilir Aliu @IlirAliu_]

Open Source Robotic Arm for All Developers
[📍Github Below]

A robotic arm project (reBot-DevArm) dedicated to lowering the barrier to learning Embodied AI.

They focus on "True Open Source" (not just the code), they unreservedly open source everything:

> Hardware Blueprints: Source files for sheet metal parts and 3D printed parts.
> BOM List: Detailed down to the specifications and purchase links for every single screw.
> Software & Algorithms: Python SDK, ROS1/2, Isaac Sim, LeRobot, etc.

Credit to the Seed Studio and thanks for reaching out, Elaine Wu!

📍GitHub: https://t.co/2YsUMBlS4s

—-

Weekly robotics and AI insights.
Subscribe free: https://t.co/9Nm01QUcw3

[Mustafa @oprydai]

think robotics is just motors?

• control loops → differential equations + real-time systems
• state estimation → bayesian inference + stochastic processes
• slam → graph optimization + nonlinear least squares
• path planning → search algorithms + computational geometry
• motion planning → kinematics + constrained optimization
• perception → computer vision + representation learning
• sensor fusion → probabilistic modeling
• manipulation → dynamics + contact physics
• localization → filtering + map consistency
• trajectory tracking → control theory
• embedded integration → hardware-software co-design
• actuation → electromechanics + power electronics
• fault handling → redundancy + safety engineering

robotics is a stack of hard problems glued together.

[Massimo @Rainmaker1973]

«The evolution of dance»

[📹 iambob_33]

[Aaron Slodov @aphysicist]

quick history lesson for those learning about the current state of manufacturing: companies used to design and manufacture their products and be vertically integrated (1800 to 1950), then we convinced ourselves that manufacturing was too expensive. we decoupled everything, shifted to free trade and services, while the complexity of the built world exploded from new science and engineering. since actual manufacturing was deemed the lowest value work, we offshored everything deemed unnecessary until only the highest margin activities remained. we gave up the essential building blocks of industrial society and ignored the existential risk.

we’re still the second largest manufacturing base globally by “output” but that’s meaningless when nobody knows how to machine anything anymore. those numbers measure dollars not capability, just financial engineering and automation running the last functioning factories, fumes from an engine built decades ago. we can pump out aerospace parts in existing facilities, but try to stand up a new production line for something basic? try to get custom tooling made? try to find someone who can troubleshoot when the cnc does something weird? good luck. the real metric should be something like “can we make new types of things we’ve never made before” or “how fast can we go from design to production” or “do we have anyone under 40 who can read a blueprint and cut metal.” by those measures we’re failing catastrophically. we’re like a rich kid who inherited a ferrari but can’t change the oil.

we’ve hemorrhaged tribal knowledge from decades of devaluing trades. when your tool breaks, there’s three guys left who know how to fix it and they’re all 65. our hubris towards “lower value” work has come full circle: we can’t make what we need, when we need it. we’ve become purely reactive, only responding when crisis hits. we optimize for quarterly profits, not our people. we’ve gone soft on hard problems. meanwhile, adversarial nations control our critical supply chains and can weaponize them at will. globalization’s utopian promise crashes against the reality that we haven’t transcended human nature or geopolitical competition.

but here’s the thing: the macro is shifting and new technology is handing us a generational opportunity. supply chain fragility is now undeniable. capital is flowing back into hard tech. ai and automation can compress decades of learning curves. software defined manufacturing can make one factory as flexible as ten. a new generation actually wants to build things again, not just apps. we don’t need to beat china at their own game, we just need to make sure they aren’t the only game in town. the next decade will determine whether america remains capable of building its own future or becomes a cautionary tale about empires that forgot how to make things. the tools are here. the capital is here. the urgency is here. we just need the will to get to work.

[Brendon 🤘 @brendonbuilds]

coming soon to @github

[Circuit Basics @circuitbasics]

Best Practices for Designing a PCB Layout
#circuitbasics #electronics #arduino #raspberrypi https://t.co/oGBujc2Now

[Tτ @teraytech]

Microstrip filters project PCB and test data files are now accessible on my github. Have fun😇https://t.co/DwjmTD0xDi

[Towaki Takikawa / 瀧川永遠希 @yongyuanxi]

Dataset of 100k STL files and G-code for 3D printing

[🆂🆄🅻🅵🆄🆁🅾🅸🅳 @sulfuroid]

My hydroponic system control board v2 is built. pH check. Water flow check. Nutriment concentration chech. Powers leds, starts water pump, corrects pH with dispenser, adds nutriments.. etc ...

[Fang @FangSystems]

Here's a side project ive been working on:
An absurdly cheap and easy manufacture jet engine for long range UAVs. ($500, 10N thrust).

I tried to do this back in highschool but now with the power of ANSYS, CF Turbo and JLCPCB its going to work this time.

[Juanbe @juanbebg]

@DVonSchwarz https://t.co/7aLs3R9WpO

[Ilir Aliu @IlirAliu_]

You don’t need a lab to give your robot touch. 🖨️ Just a printer.

[Paper & code ⬇️]

eFlesh is a 3D-printable, customizable tactile sensor built for makers and researchers.

No fancy lab setup, no expensive gear… just a $5 bill of materials, a hobbyist printer, and a simple online tool.

Why eFlesh is a leap forward
✅ Shape, size, and stiffness are fully customizable for your use case
✅ Simple setup with neodymium magnets and a magnetometer PCB
✅ Low magnetic interference and no cross-talk across sensor tiles
✅ Strong performance with 95 percent slip detection accuracy and sub-mm force resolution

Thank you for sharing, @Raunaqmb!

Make your own: https://t.co/6h7wG099ft

[Ilir Aliu @IlirAliu_]

1.7× faster inverse kinematics in pure Python, GPU-accelerated!
Fully open-source and built for scale
[📍Bookmark for later]

PyRoki, a modular toolkit for robot kinematic optimization, supporting
- inverse kinematics
- trajectory optimization, and
- motion retargeting.

Built for flexibility, but also fast:
✅ Runs on CPU, GPU, and TPU
✅ Outperforms cuRobo in speed, success rate, and accuracy
✅ Fully in Python, easy to integrate and extend

Whether you’re working with industrial arms, simulation environments, or humanoids, PyRoki is designed to fit your stack and scale with your hardware.

Thank you, @ChungMinKim, for sharing!!

Website: https://t.co/YRlgDbTJ0Y
Code: https://t.co/8D1pcE0oxq
Paper: https://t.co/I6rqtyMWWP

This one’s for researchers, builders, and anyone tired of debugging their IK stack.