Robotics and Biology Laboratory

News

Laborbesuch von Prof. Michael Beetz

Wir freuen uns immer über Besuch von Kollegen.

Professor Michael Beetz Leiter des Instituts für Künstliche Intelligenz (IAI), besuchte heute unser Labor.

Dabei hatte Adrian die Gelegenheit, ihm die In-Hand-Manipulationsfähigkeiten unserer RBO Hand 3 zu zeigen.

 

Willkommen Alexander!

Wir begrüßen unser neues RBO-Mitglied Alexander Koenig!

Wir begrüßen unser neues RBO-Mitglied Alexander Koenig!

Alexander wird sich im Rahmen seines Promotionsstudiums mit den algorithmischen Grundlagen der verkörperten künstlichen Intelligenz beschäftigen.

New IROS Paper "Augmentation Enables One-Shot Generalization In Learning From Demonstration for Contact-Rich Manipulation"

Rethinking robot learning from demonstration: Can demonstrations alone teach robots complex manipulation skills? Our latest paper shows that with augmentation, one-shot learning from demonstration is possible! Discover how the extracted policies generalize across various objects in 'Augmentation Enables One-Shot Generalization In Learning From…

Paper "Dexterous Soft Hands Linearize Feedback-Control for In-Hand Manipulation" presented on IROS 2023

Have you ever thought about representations of soft hands for in-hand manipulation? Or about feedback controllers that are suitable for compliant systems? If so, check out our new IROS23 paper.

Have you ever thought about representations of soft hands for in-hand manipulation? Or about feedback controllers that are suitable for compliant systems? If so, check out our new IROS23 paper.

This paper presents a feedback-control framework for in-hand manipulation (IHM) with dexterous soft hands that enables the acquisition of manipulation…

Paper "In-Hand Cube Reconfiguration: Simplified" presented on IROS 2023

In this work we present a simple approach to in-hand cube reconfiguration with gravity being the only source of actuation.

We present a simple approach to in-hand cube reconfiguration. By simplifying planning, control, and perception as much as possible, while maintaining robust and general performance, we gain insights into the inherent complexity of in-hand cube reconfiguration.

For more details have a look at our paper or visit our project website and YouTube…

Adrian gave a Talk at Merantix Momentum

Adrian Sieler presented his research on "The Interplay of Learning and Embodiment for Dexterous Robotic Manipulation" at Merantix Momentum

Adrian Sieler presented his research on "The Interplay of Learning and Embodiment for Dexterous Robotic Manipulation" at Merantix Momentum 🤖📷 #DexterousManipulation #MorphologicalComputation

The recording can be found here.

New AlphaLink Extension Preprint

Our new preprint (with RappsilberLab) which extends AlphaLink to protein complexes is out!

Our new preprint (with RappsilberLab) which extends AlphaLink to protein complexes is out!
We show that directly integrating crosslinking MS data into AlphaFold-Multimer can substantially improve the prediction quality!

AlphaLink Nature Paper Published

Our AlphaLink paper (with the RappsilberLab) on integrating crosslinking MS data into AlphaFold2 is out!

Our AlphaLink paper  (with RappsilberLab) on integrating crosslinking MS data into AlphaFold2 is out!
We show that crosslinks can substantially improve the prediction quality of challenging targets and steer predictions towards in situ conformations! Great work Kolja Stahl!

Welcome Paul

We welcome our new RBO member Pu (Paul) Xu.

We welcome our new RBO member Pu (Paul) Xu. Paul will work on engineering paradigms for intelligent robotic systems as part of SCIoI.

Welcome Duc

We welcome our new RBO member Huu Duc Nguyen.

We welcome our new RBO member Huu Duc Nguyen. Duc will work on the SCIoI Project 'Intelligent Kinematic Problem Solving' as part of his doctoral studies.