IROSA: Interactive Robot Skill Adaptation using Natural Language Foundation models have demonstrated impressive capabilities across diverse domains, while imitation learning provides principled methods for robot skill adaptation from limited data. Combining these a...

Check the paper and let me know when you are having any questions: arxiv.org/abs/2603.03897

IROSA: Interactive Robot Skill Adaptation Using Natural Language Foundation models have demonstrated impressive capabilities across diverse domains, while imitation learning provides principled methods for robot skill adaptation from limited data. Combining these a...

IEEE Explore Paper link: doi.org/10.1109/LRA....
ArXiv paper link: arxiv.org/abs/2603.03897

🚀 New paper in IEEE RA-L!

IROSA: Interactive Robot Skill Adaptation using Natural Language

Tell a robot "go faster" or "there's an obstacle" - it adapts. The LLM selects validated tool functions, no direct robot control: a zero-shot, local LLM.
#robotics #LLM #IROS2026 #HRI
👇

Interactive incremental learning of generalizable skills with local trajectory modulation The problem of generalization in learning from demonstration (LfD) has received considerable attention over the years, particularly within the context of movement primitives, where a number of approac...

arxiv.org/abs/2409.05655

Arrived at IROS 2025 to present our work - Interactive incremental learning of generalizable skills with local trajectory modulation. 🦾

Feel free to pass by to see my talk and chat with me at my poster. 💬

Robotic Imitation Learning 4
🗓️ Wed, Oct 22· 5:10PM
📍 Room 311 B

Link to paper in comments.

Got some recent research related to 6D Object Pose Estimation? Want to present it at #ICCV2025 in Hawaii? 🌴

Then submit and present at the 𝟏𝟎𝐭𝐡 𝐈𝐧𝐭𝐞𝐫𝐧𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐖𝐨𝐫𝐤𝐬𝐡𝐨𝐩 𝐨𝐧 𝐑𝐞𝐜𝐨𝐯𝐞𝐫𝐢𝐧𝐠 𝟔𝐃 𝐎𝐛𝐣𝐞𝐜𝐭 𝐏𝐨𝐬𝐞 (𝐑𝟔𝐃).

Paper deadlines: Jun 30 (in-proceedings), Aug 29 (non-proceedings)

Paper link: elib.dlr.de/214144/

Checkout our paper at #ICRA2025 in Atlanta 🇺🇸

How can robotic systems articulate truthful answers about their internal knowledge in natural language? 🤖

📅 Presentation: May 20th, 2025, at 16:45 EDT in Room A314
📍 Poster session: Immediately following the talks

Queer in Robotics Purpose

Hello, Bluesky! 🌈🤖 Thrilled to introduce Queer in Robotics! We're building a welcoming, safe space for queer roboticists and shining a light on queer issues in robotics. Join us! 💫
sites.google.com/view/queerin...

Getting SAC to Work on a Massive Parallel Simulator: An RL Journey With Off-Policy Algorithms (Part I) | Antonin Raffin | Homepage This post details how I managed to get the Soft-Actor Critic (SAC) and other off-policy reinforcement learning algorithms to work on massively parallel simulators (think Isaac Sim with thousands of ro...

"As researchers, we tend to publish only positive results, but I think a lot of valuable insights are lost in our unpublished failures."

New blog post: Getting SAC to Work on a Massive Parallel Simulator (part I)

araffin.github.io/post/sac-mas...

A screenshot of output from the blenderproc tool, with associated options. Left: Mesh Middle: Point Cloud Right: Depth

Publication-ready visualization of 3D objects and point clouds in seconds, using @blender.org and BlenderProc.

hummat.github.io/bproc-pubvis/

GitHub: github.com/DLR-RM/inter...

ArXiv Paper: arxiv.org/abs/2409.05655

IEEE Explore: ieeexplore.ieee.org/document/108...

YouTube: www.youtube.com/watch?v=nqig...

🚀 Exciting news! Our work has been published in IEEE 𝘙𝘰𝘣𝘰𝘵𝘪𝘤𝘴 𝘢𝘯𝘥 𝘈𝘶𝘵𝘰𝘮𝘢𝘵𝘪𝘰𝘯 𝘓𝘦𝘵𝘵𝘦𝘳𝘴 (𝘙𝘈-𝘓). 🌟

📄 𝗧𝗶𝘁𝗹𝗲: 𝘐𝘯𝘵𝘦𝘳𝘢𝘤𝘵𝘪𝘷𝘦 𝘪𝘯𝘤𝘳𝘦𝘮𝘦𝘯𝘵𝘢𝘭 𝘭𝘦𝘢𝘳𝘯𝘪𝘯𝘨 𝘰𝘧 𝘨𝘦𝘯𝘦𝘳𝘢𝘭𝘪𝘻𝘢𝘣𝘭𝘦 𝘴𝘬𝘪𝘭𝘭𝘴 𝘸𝘪𝘵𝘩 𝘭𝘰𝘤𝘢𝘭 𝘵𝘳𝘢𝘫𝘦𝘤𝘵𝘰𝘳𝘺 𝘮𝘰𝘥𝘶𝘭𝘢𝘵𝘪𝘰𝘯

Links in comments.

#robotics #research #interactiveLearning #imitationLearning #machineLearning #IEEE

GitHub - DLR-RM/BlenderProc: A procedural Blender pipeline for photorealistic training image generation A procedural Blender pipeline for photorealistic training image generation - DLR-RM/BlenderProc

github.com/DLR-RM/Blend...

Markus Knauer on LinkedIn: #deeplearning #datageneration #blender In need of data for your deep learning approach? Than check this out: pip install blenderproc #deeplearning #datageneration #blender

www.linkedin.com/posts/markus...

Paper: elib.dlr.de/205203/1/Bus...

Workshop Website: sites.google.com/view/gai-hri...

Markus Knauer on LinkedIn: #rss2024 #rss #delft #robotlearning #foundationmodels I am going to present our work at the GenAI-HRI Workshop today at Robotics: Science and Systems 2024 Conference #RSS2024 in Delft! Title: Grounding Embodied…

www.linkedin.com/posts/markus...

Interactive incremental learning of generalizable skills with local... The problem of generalization in learning from demonstration (LfD) has received considerable attention over the years, particularly within the context of movement primitives, where two prominent...

openreview.net/forum?id=PAg...

Interactive incremental learning of generalizable skills with local trajectory modulation The problem of generalization in learning from demonstration (LfD) has received considerable attention over the years, particularly within the context of movement primitives, where a number of approac...

arxiv.org/abs/2409.05655

Markus Knauer on LinkedIn: #corl2024 #conferenceonrobotlearning #machinelearning Let’s meet at #CoRL2024 in Munich where we are presenting our approach about interactive incremental learning. If you are interested see the link in the…

www.linkedin.com/posts/markus...

#ConferenceOnRobotLearning #MachineLearning #CoRL2024