Uncategorized

A big congratulations to HCRL’s student Seung Hyeon Bang for graduating with a PhD in Aerospace Engineering titled “Reactive and Predictive Whole-body Control for Agile, Robust, Versatile, and Deployable Humanoids”.

We are sharing, open source, our new design and assembly for our LEGATO Handheld Gripper for cross-embodiment robot learning. Instructions for design and assembly are here below:

https://ut-hcrl.github.io/LEGATO-Gripper

In collaboration with Dexterity Inc., we introduce a unique hardware-accelerated ray-tracing method for direct volume mesh-to-mesh discrete collision detection, which particularly excels in continuous collision detection. Kudos to Andrew Bylard, who guided this work following his groundbreaking research, and to my student Sizhe Sui, who led the technical implementation.

Link to Paper:

https://arxiv.org/abs/2409.09918v1

Introducing our latest collaborative work, named LEGATO, which describes a universal handheld smart gripper and a motion-invariant policy, that enable efficient cross-embodiment skill transfer between human users and robots with different morphologies, enhancing scalability and versatility in robotics. Congratulations to my PhD student Mingyo Seo who has led the study. In addition, many thanks to The AI Institute, Andy Park and Yuke Zhu (co-advisor of Mingyo) for their incredible contributions and support. Many thanks to Mitch Pryor for providing access to his Spot robot. Many thanks to the Office of Naval Research for supporting this project.

Pointer to paper:
https://arxiv.org/abs/2411.03682

Project Website (includes code and various videos):
https://lnkd.in/gBeW8pQZ

This comprehensive paper introduces a model for improving the grasp strength of multi-suction-cup grippers, addressing challenges in heavy object manipulation. It presents new constraints for trajectory planning and optimization, solves load distribution issues with a quadratic program, and validates the model through experimental results.

Our team, UT-SOTON, is leading the entry into the XPrize Wildfire competition in collaboration with the Jackson School of Geosciences, the University of Southampton, and TU Delft. The following news are from Austin’s KXAN.

Check out this video reel of the projects of Prof. Sentis’ new graduate class Design of Human Centered Robots, taught this past Spring semester. Every robot and system shown in this video is built from scratch. We are very impressed by the students creativity and skills. The arm, reinforced with carbon fiber, includes custom-built cycloidal gears inspired by Haddington Dynamics. And the dexterous hand? Although it’s a replica of an open source design, the 3D printing and assembly by the students are absolutely mind-boggling! Hand blueprints by https://robotnanohand.com, arm design inspired by Haddington Dynamics.

We had a great time at ICRA 2024. Jeeeun Lee presented a very nice paper on Jerk Constrained Time Optimal Path Planning and Control:

We also planned activities for our IEEE Technical Committee on Whole-Body Control:

And finally, current and past HCRL students got together in Yokohama during the conference:

We have launched a new website with all results of our NSF Multidisciplinary project. Click below!

Want to understand how humans react to quadruped robots? We’ve published a dataset of human-subject data, including stress responses (physiological measures + interviews). Click on the image below to access the dataset.