Weekly Robotics #319

Issue 319

Last week, was the 12th Automatica I did not attend. Every year, I promise myself I'll attend, but then life catches-up to me. Hopefully I'll make it to the next one. I was trying to find some coverage of the interesting demonstrators online, and liked this video reel by the Manufacturing Millennial. Have you found any other summaries of this event that caught your attention?

Digital Robotics Archive Collections

Digital Robotics Archive Collections cover

I just came across this treasure trove of cool robotics projects that is maintained by CMU, and I found it to be a great rabbit hole to fall into. Happy browsing!

CMU has launched The Robotics Project: a unique partnership between the University Libraries and the School of Computer Science to create a home for the past, present, and future of robotics. We’re working to create the pre-eminent archival, historical, and instructional resource on robotics as a scientific discipline, a field of practice, and a cultural force.

Source

Gemini Robotics On-Device brings AI to local robotic devices

DeepMind team announced an On-Device VLA model that’s optimized to run efficiently on robots. The model was made for bi-arm manipulation. I recommend checking-out the videos showing this VLA model on the linked website.

Source

OpenExo - Open-Source Exoskeleton

OpenExo - Open-Source Exoskeleton cover

If you would like to build your own exoskeleton, look no further! The design and software for this work by the researchers from Northern Arizona University. To learn more about this project, check out this paper. Unfortunately, I didn’t find any indication of how much a complete system could cost.

Source

A Versatile Quaternion-based Constrained Rigid Body Dynamics

A Versatile Quaternion-based Constrained Rigid Body Dynamics cover

We present a constrained Rigid Body Dynamics (RBD) that guarantees satisfaction of kinematic constraints, enabling direct simulation of complex mechanical systems with arbitrary kinematic structures. To ensure constraint satisfaction, we use an implicit integration scheme. For this purpose, we derive compatible dynamic equations expressed through the quaternion time derivative, adopting an additive approach to quaternion updates instead of a multiplicative one, while enforcing quaternion unit-length as a constraint.

Source

Learning Accurate Throwing with High-frequency Residual Policy and Pullback Tube Acceleration

Learning Accurate Throwing with High-frequency Residual Policy and Pullback Tube Acceleration cover

Very cool research from ETH, where an ANYmal robot was taught to throw objects using its whole body and a robot manipulator attached to its back.

Source

Can MechE’s Survive the AI Revolution? I Asked a Boston Dynamics ML Engineer (ex-MechE)

In this video, Leon interviews Jesse Miller, a senior reinforcement learning engineer at Boston Dynamics. In the interview, you will learn how a classical control engineer can end up doing reinforcement learning and get some tips how to enter the AI field.

Source

Events

For more robotic events, check out our event page.

Want to promote your product or service in Weekly Robotics? Check out our advertising options.