-T.K.-
@t_k_233
System architecture design for humanoid robotics
少し前から知り合いとプロジェクトを立ち上げて進めていたのですが、諸々の事情で一旦見直しに... これはその試作機、リンク機構で3自由度で結構いい感じなんだけど厳しい要素もあったりでどのみち没案になりそうなので公開
Very interesting AuA
HOW DO WE TEST ACTUATORS? In this episode, we will introduce the key tests that RobStride actuators undergo, as well as the equipment and conditions we use for these tests. Through such rigorous testing, we ensure the reliability and quality of the actuators. If you have any…
This week UC Berkeley unveiled Humanoid Lite: a 1 m tall, 16 kg robot you can 3D-print at home, assemble in about a week, and walk using AI. The best bit? It’s powered by off-the-shelf parts and open-source code. 🧵
🚀 Introducing LeVERB, the first 𝗹𝗮𝘁𝗲𝗻𝘁 𝘄𝗵𝗼𝗹𝗲-𝗯𝗼𝗱𝘆 𝗵𝘂𝗺𝗮𝗻𝗼𝗶𝗱 𝗩𝗟𝗔 (upper- & lower-body), trained on sim data and zero-shot deployed. Addressing interactive tasks: navigation, sitting, locomotion with verbal instruction. 🧵 ember-lab-berkeley.github.io/LeVERB-Website/
Oh no, we are going to demonstrate our Berkeley Humanoid Lite robot at #RSS2025 today, but the leg is broken during the transportation. But worry not, because in our design, all of the structural components can be 3D printed, and glad we brought a 3D printer with us! 😆
Very cool drone shot
1️⃣/7️⃣🤖⚽ Toward Real-World Cooperative and Competitive Soccer with Quadrupedal Robot Teams Most learning-based robot-soccer work stays in simulation or tests 1v1. We field real-world games with both cooperation and competition—plus robot-human matches!
Spent three nights, finally figured out how to set up USB-CAN on Jetson with JetPack 6.0
If you guys want to take a closer look at the finger tracking in a virtual guitarist’s first-person view.
Got stuck at an intersection on the way home for two entire green light cycles due to:
Here is Claude 3's system prompt! Let me break it down 🧵
Very interesting work! A bit confused tho on how system like this can sense external torque excerted by the rest of the robot's linkage and the environment (terrain etc.)
Cycloidal Quasi-Direct Drive Actuator Designs with Learning-based Torque Estimation for Legged Robotics. arxiv.org/abs/2410.16591