Such an honor that our project is highlighted by @ERC_Research! We are so excited to put these humanoid robots into serious work powered by SIREN's breakthroughs!
Soon, we will announce open positions! We need highly motivated people with interdisciplinary background for SIREN!
I’m very happy that one of my dreams came true! Starting tomorrow I will be Assistant Professor at @CS_TUDarmstadt leading the group of intelligent robotic systems for assistance iROSA. Lots of hard work ahead, but I am grateful and excited!!
After receiving 4 offers, I am excited to announce that I have decided to keep pursuing my research goals at @TUDarmstadt as Full Professor sponsored by @Hessian_AI. I am grateful to my mentors, family, and group for their support.
#AI#Robotics
(1/2)
Congratulations to Georgia Chalvatzaki @GeorgiaChal! Last year the #AI expert was appointed as Assistant Professor, today #TUDa-President Tanja Brühl presented her with the certificate of appointment as Professor at the department of computer science @CS_TUDarmstadt @Hessian_AI
ALT TUDa-President Professor Tanja Brühl presents Professor Georgia Chalvatzaki with the certificate of appointment.
My favourite ML book arrived just in time for the spring semester!!! Very excited that this semester I will be teaching two courses: Statistical Machine Learning and Reinforcement Learning @CS_TUDarmstadt @TUDarmstadt Additional hands-on robotics seminar on the way. Stay tuned!
I'm so excited that 3 of my papers got accepted in #ICRA2021!!! Thank you to all my co-authors! Now we push hard for #RSS2021 and #IROS2021 next week! Stay strong! Stay positive!🙏🙏🙏💪💪💪
First in-person lecture and first lecture as Professor @CS_TUDarmstadt teaching first lecture on Statistical Machine Learning :)
I am so happy to see people in the university again! I enjoyed it so much and cannot wait for the next lectures to come!
The message should be heard, both by ML industry and some research groups: For solving robots you need to understand the physical system, its potential and limitations, know your Math, Physics, Control, and use ML when needed in the loop with classical methods!
#Robotics
I am happy to announce that I am expecting applications for PhD students through the #EllisPhD program! If you are interested in Interactive Robot Perception & Learning apply till November 15!
#ELLIS#robotlearning#machinelearning#AI
Great news! We got a @HorizonEU proposal accepted on Advancing the physical intelligence and performance of #robots towards human-like bi-manual objects manipulation! Along with a great consortium from 🇬🇷🇩🇪🇮🇹🇪🇸🇦🇹🇬🇧🇵🇱More Ph.D. positions open soon! @CS_TUDarmstadt @Hessian_AI
I am delighted that we got 6 papers accepted @2022Iros#IROS22 at Kyoto! Well done to our students @CS_TUDarmstadt and our collaborators for this success! 🧵 coming with a brief reference to our papers 👇
Honored to give an Early Career Keynote at #CoRL2024! My talk, "On the Quest for Robotic Embodied Intelligence," will explore how structure in robot learning enables faster, safer, and adaptive robot behaviors. Excited to share this vision! 🚀
corl.org/program/speakers#robotics
Honored for the early career keynote at #CoRL2024 last week—unforgettable milestone! Grateful to the community for the recognition. From a small lab in Greece to leading my own group @TUDarmstadt, this journey fuels me to push #RobotLearning further. #ERCStG SIREN is the start!🦾
My group @CS_TUDarmstadt has several open #PhD positions on a variety of topics encompassing Interactive #Robot Perception and Learning for Mobile Manipulation Systems -- positions funded by @dfg_public and @HorizonEU
Apply now and join us! irosalab.com/open-positions/
ALT Active perceptive mobile manipulation with TIAGo++
I am thrilled to receive the award of AI Newcomer 2021, among all these great researchers in the diverse field of AΙ research!
A great thanks to the people who supported us with their votes, to the jury, and the KIcamp from @informatikradar
Congratulations: AI Newcomer 2021 of the category technical and engineering science is Dr. Georgia Chalvatzaki! 👏
"AI advances robotic research for developing intelligent agents that interact safely and assist humans" - Georgia Chalvatzaki
#KICamp21@GeorgiaChal
Thrilled to be at #ICML2023 in Hawaii 🌺 presenting MINT: Maximum Information Keypoints, an entropy-driven unsupervised representation learning approach that treats keypoints as Info transmitters which allow for consistent representation of dynamic scenes in videos. (1/4)
ALT Come and go scenario showcasing the ability of MINT to dynamically adjust the number of keypoints to preserve spatiotemporal in the video demonstration
I am excited to announce that I got the highly prestigious Emmy Noether #DFG for AI grant to start my research group iROSA as PI for Robot Learning of Mobile Manipulation @TUDarmstadt I am grateful for my mentors @Jan_R_Peters and Oskar von Stryk, as well as my department support
🚨Beyond happy and deeply honored to share that my project, Structured Interactive Perception & Learning for Holistic #Robotic Embodied #Intelligence (SIREN), has been awarded an #ERC Starting Grant! 🎉🤖🧜♀️A massive thank you to the @ERC_Research for believing in my vision (1/3)
We are honored to be nominated for the Best Paper Award on Mobile Manipulation for the second year in a row #IROS2023. Congratulations to the first co-authors of our paper @n_w_funk and Luca Lach. Also congratulations to the winners!
Heading to #RSS2024 at Delft. Excited to meet the RSS community and present some exciting research from the PEARL group and collaborators 🤖
(1/8) On Monday, July 15, we will be co-organizing the GenAI-HRI workshop sites.google.com/view/gai-hr…
What amazes me the most is how far you can go with 2-fingered grippers even if teleoperated. Makes you wonder if more dexterity is needed for a large range of tasks in homes. Would be interesting to transfer to our bimanual mobile robots.
Mobile ALOHA's hardware is very capable. We brought it home yesterday and tried more tasks! It can:
- do laundry👔👖
- self-charge⚡️
- use a vacuum
- water plants🌳
- load and unload a dishwasher
- use a coffee machine☕️
- obtain drinks from the fridge and open a beer🍺
- open doors🚪
- play with pets🐱
- throw away trash
- turn on/off a lamp💡
Project website: mobile-aloha.github.io/
Co-lead @tonyzzhao, advised by @chelseabfinn
(amazing photographing from @qingqing_zhao_ )
I am deeply honored and excited to deliver a keynote speech about our research and what we aim for at the iROSAlab.com @CS_TUDarmstadt at the upcoming #IROS2022 in Kyoto! I hope we can all meet in person for this top-tier robotics event!
I am heading to Japan to represent @CS_TUDarmstadt @TUDarmstadt@Hessian_AI at #IROS2022 in Kyoto, together with an amazing team from the iROSA and @ias_tudarmstadt groups!
For me this is the most exciting conference so far! See in the following thread why👇
(1/5)👇
I am delighted to share that I was featured in Woman in Computer Vision on Computer Vision News @RSIPvision. It was great chatting with Ralph! Thank you for allowing me to discuss the challenges and opportunities for interactive robot perception and action in robot learning!
👇
How can we reason about task sequences to ensure feasibility in a single pipeline that scales to complex #robot manipulations? In our new work sites.google.com/view/rl-mee…, we propose a novel hierarchical framework that combines MILP with #GNN#RL and Grasp and Motion planning.
I am Detroit, attending @ieeeiros, and I am excited about all the activities coming up for me and my collaborators!
(1/n) Tomorrow, we will host a half-day workshop on Policy Learning on Geometric spaces, the challenges and opportunities
I strongly believe this work is revolutionary for trajectory optimization! Congratulations to @an_thai_le for delivering this amazing research! I am honoured to have contributed to this work! See you at #NeurIPS2023
I am very happy to announce that our work: "Accelerating Motion Planning via Optimal Transport" has been accepted to #NeurIPS2023!
Project website: sites.google.com/view/sinkho…
If you are at #ICLR2024 and are interested in Sim2Real for RL, reach out to @GabrieleTiboni1 to talk about our work DORAEMON that formulates the appropriate optimization problem over randomizing dynamics parameters to ensure policies become robust yet adaptive in the real world.
Excited to announce our interdisciplinary #ICRA2023 Workshop on effective Representations, Abstractions, and Priors for Robot Learning (#RAP4Robots) on May 29! Info and call for papers at: sites.google.com/view/rap4ro…
Deadline for abstracts: 24 March
@ieee_ras_icra
(1/3)
Smooth, differentiable constraints are fundamental for enabling robots to plan their actions and act safely. In our new work irosalab.com/redsdf/, we derive Regularized Deep Signed Distance Fields (ReDSDF) (1/6)
#Robots#DeepImplicitFunction#DeepLearning
🎉 Celebrating the International Day of Women and Girls in Science 🚀
Today, we honor the trailblazing #WomenInScience who are shaping the future. Their innovation, strength, and determination inspire us all. Let’s continue to break barriers!
#WomenInSTEM#WomenInRobotics
ALT Image generated via Dall-E representing women in robotics
Honoured to be acknowledged as top reviewer for #NeurIPS2023. Thank you to the AC that proposed me and the PC for their work and the free registration ticket :) neurips.cc/Conferences/2023/…
🤖📝 Our new research explores how to finetune and ground #GPT2 on scene graphs for autonomous robot reasoning and long-horizon task planning. Our work contributes to the grand quest for general #AI#Robotics
🧵
(1/7)
Looking at the #teslabot and as a roboticist, I am impressed by what the engineers achieved in a year for this prototype. However, the behaviours demonstrated are less impressive than that of Honda’s #Asimo from twenty years ago. Read the analysis of @chubicki (1/2)
Humbled to be included in this list curated by @ben_eysenbach! Please consider inviting more women+ to your planned events. Recognize, retain, and attract more diverse talent!
If you are interested in doing a PhD with us @CS_TUDarmstadt apply now using the links in the comments. We have open positions on a range of topics spanning representation learning, reasoning & geometric robot learning for mobile manipulation.
#robotics#AI#PhDpositions
👇
Excited to participate at the 40th @ieee_ras_icra in London! We will be around organizing events and presenting our work from the collaborations of my group and @ias_tudarmstadt @CS_TUDarmstadt.
See the thread about our contributions/events:
👇
Very proud of our work getting in #ICRA2024. We propose a new pipeline for active mobile manipulation in unknown cluttered scenes.
Kudos to @SophieLueth and @SnehalJauhri for the great work and real robot experiments 🤖🦾
Check it out 👇
It was an amazing trip to Japan! We had a great series of presentations @2022Iros and then a series of research visits in ATR, Uni Osaka, NAIST, RIKEN AIP, university of Tokyo. Here is a photo of our last night in Tokyo @CS_TUDarmstadt @Hessian_AI
📣 Introducing DDiffPG!
Check out our new paper for learning #Diffusion Policies with online #ReinforcementLearning ensuring that we discover and maintain #multimodal behaviors!
👇
How can an RL agent successfully solve a task while showcasing versatility of behaviors—a property intuitive to intelligent systems like humans?
Introducing Deep Diffusion Policy Gradient (DDiffPG) - our new algorithm for learning multimodal behaviors from scratch! (1/5)
@theCamusean, @n_w_funk, @Jan_R_Peters and me are very happy to receive the best paper award at the workshop on Geometric Representations: The Roles of Screw Theory, Lie algebra, & Geometric Algebra @ieee_ras_icra! Thank you to the organizing committee!
Our RA-L + #IROS2022 paper "Robot Learning of Mobile Manipulation with Reachability Behavior Priors" has been selected as a finalist of the
IROS Best Paper Award on Mobile Manipulation!!!
Congrats to my Ph.D. student
@SnehalJauhri
for the great work!
@CS_TUDarmstadt
We had a blast @RoboticsSciSys#RSS2024! Congrats to organizers for an amazing conference. RSS gives you the opportunity to get a good overview of the field thanks to its single-track nature. On top, workshops are great forums of discussions and exchange. See you in LA next year!
Honored to give an Early Career Keynote at #CoRL2024! My talk, "On the Quest for Robotic Embodied Intelligence," will explore how structure in robot learning enables faster, safer, and adaptive robot behaviors. Excited to share this vision! 🚀
corl.org/program/speakers#robotics
Full house at the Scaling Robot Learning Workshop at #RSS2022 Very interesting insights from the first speakers! I’m excited to share my insights on the topic this afternoon!
The future is about intelligent dexterous mobile manipulation robots that watch understand and do! I cannot agree more! Amazing plenary by Marc Raibert @BostonDynamics@ieee_ras_icra
I'm very happy that our paper "Deep Leg Tracking by Detection and Gait Analysis in 2D Range Data for Intelligent Robotic Assistants" got accepted at #IROS2021
Great news this morning! Our paper “Monte-Carlo Robot Path Planning” was accepted both to RAL and IROS! Congrats to our student @tuanquangdam for this success! This paper introduces an algorithmic framework for robot path planning under uncertainty with theoretical guarantees!
I am delighted to offer an advanced seminar for M.Sc. students @CS_TUDarmstadt on Intelligent Robotic Manipulation this winter semester, accompanied by an exciting Project Lab.
Brief description in 🧵
#Robotics#academia
Excited to kick-off the #ReinforcementLearning lecture today @CS_TUDarmstadt. I look forward to the interactions with the new curious students that are always rewarding and make us gain deeper knowledge of the field ;) See you all in class 🤖
The third edition of the Reinforcement Learning course at @CS_TUDarmstadt started today, with a very nice introduction by Prof. @GeorgiaChal !
I'm excited to teach some advanced Deep RL methods again this year!
Very happy that our paper with @n_w_funk@_bbelousov@Jan_R_Peters "Learn2Assemble with Structured Representations and Search for Robotic Architectural Construction," was accepted at #CoRL2021
Congratulations to the Best Paper awardees of the #RAP4Robot workshop @ieee_ras_icra. A great thanks to all attendees and amazing speakers for the insightful discussions, and a special thanks to our sponsor @nvidia for the amazing gifts!
I am very excited to co-organize this workshop. Geometry plays a key role in robotics and we need to find new ways of integrating geometric representations in robot learning! Looking forward to our discussions at #IROS2023!
On this #InternationalWomensDay, let’s remember that progress in Robotics & AI thrives on diversity. Proud to support and be part of a community that values and uplifts women's contributions every day. Let’s keep pushing the boundaries together. #WomenInTech#WomenEmpowerment
Still looking for motivated people to join my group at @CS_TUDarmstadt @Hessian_AI to research Interactive Robot Perception & Learning. If you are at #IROS2023 and you are interested to know more about our research, ping me!
My group @CS_TUDarmstadt has several open #PhD positions on a variety of topics encompassing Interactive #Robot Perception and Learning for Mobile Manipulation Systems -- positions funded by @dfg_public and @HorizonEU
Apply now and join us! irosalab.com/open-positions/
ALT Active perceptive mobile manipulation with TIAGo++
The @ieeeras Women in Engineering committee launches a new Travel support for the major RAS conferences, that refers to underrepresented groups and minorities who want to enter the field! No paper required!
First info for @ieee_ras_icra: ieee-ras.org/about-ras/lates…
Pls rt!
We are excited to announce this year’s edition of the RSS Pioneers workshop > sites.google.com/view/rsspio…
Applications are due March 22, 2021 (AoE). We are planning this virtual event to be as interactive as possible to transfer a full experience! Stay tuned for more!
Today students of @CS_TUDarmstadt showed their Sense-Plan-Act solutions for sequential cube stacking with TIAGo++ at the #iROSA lab. It was a fun and educating experience — a glimpse into the life of a robotics engineer.
Lots of bloopers but will keep those private 😅
DDiffPG, our algorithm for learning multimodal diffusion policies with online #ReinforcementLearning is accepted to @NeurIPSConf! Congrats to @softraeh, who led this work during his master's (soon to start his PhD at PEARL @CS_TUDarmstadt). A hearty congrats to our co-authors!
How can an RL agent successfully solve a task while showcasing versatility of behaviors—a property intuitive to intelligent systems like humans?
Introducing Deep Diffusion Policy Gradient (DDiffPG) - our new algorithm for learning multimodal behaviors from scratch! (1/5)
People @ieee_ras_icra join us at our workshop on effective Representations, Abstractions and Priors for Robot Learning this Monday! We have a fantastic lineup of speakers, intriguing panel discussions, amazing contributions and incredible awards!
Do not miss #RAP4Robots👇
ALT RAP4Robots workshop at ICRA23 May 29 Excel London
The iROSA group @CS_TUDarmstadt @Hessian_AI offers multiple Ph.D. positions in the following exciting topics:
- Geometric and Physics-Aware #RobotLearning
- Lifelong Explainable #RobotLearning with pretrained multimodal models (co-supervision with @CarloDeramo)
(1/3)
I am so grateful to belong to this family!!! Mostly grateful for this year that changed my life and so many new things started, especially in these difficult times, and I cannot wait for the exciting new year to come! #irosa@ias_tudarmstadt#Christmas2021#vaccinated
If you are attending #RLDM2022 check out today our poster on #ReinforcementLearning for Robot Assembly. 🤖🧱
@n_w_funk will be there to tell you all about our work on bringing GNN-RL to Long-horizon Robot Decision-Making!
I am excited to be @ieee_ras_icra in Yokohama, Japan with @SnehalJauhri@SophieLueth from the PEARL group @CS_TUDarmstadt @Hessian_AI to present new exciting work on learning for mobile manipulation!
Catch us on the following events:
I am honored to join together with @RobobertoMM, the @ieeeras technical committee of Mobile Manipulation, alongside @correlllab@Ken_Harada and Mehmet Dogar. I am very excited to serve the TC and organize many activities on mobile manipulation! Stay tuned!
Very important to rejected #ICRA2021 papers!!! #IROS2021
Authors of a paper that has just been rejected by ICRA 2021 can provide, optionally, review info of the (possibly revised) resubmission. A single PDF file with (1) cover letter (2) full reviews from ICRA 2021 and (1/n)
Honoured to serve as Workshops Chair for #CoRL2024.
To help make our workshops program more diverse, please take a note on the new requirements in our Call for Workshops.
Do not hesitate to contact me in case of questions.
We look forward to your proposals!
It was a pleasure to visit and talk @BristolRobotLab. Impressive facility and great researchers hosted there! Thank you @efi_psomopoulou and @nathanlepora for organizing this symposium! I look forward to future interactions!
Incredibly proud of my friend and amazing scientist Kostas Alexis @arlteam for his keynote at #IROS2023 showcasing the inspiring work of Team Cerberus and their ongoing work on robotic resilience!
Overly excited heading to London for #CoRL2021 to present our work and to meet colleagues and friends again after two years! I cannot wait for all the exciting works and discussions. Let’s catch up!
The Conference on #Robot#Learning 2021 will be held on Nov 8-11 in London, UK & virtually. Exciting new changes, including:
- both regular and blue sky paper submission types, and
- use of #openreview
June 18th is the submission date!
Find out more: robot-learning.org/
I’m in NY to attend @RoboticsSciSys where I will be giving two invited talks. On Monday at the workshop on Scalable Robot Learning and on Friday at the workshop on @SIHRworkshop People attending #RSS2022 I’ll be happy to catch up!
Very pleased that our work "Placing by Touching: An empirical study on the importance of tactile sensing for precise object placing" with Luca Lach, @n_w_funk, @skadge, Helge Ritter, and @Jan_R_Peters got accepted at #IROS2023!
Check out our new Survey in Deep Generative Models for Robotics! We discuss different aspects of the application of DGMs and investigate how to improve generalisation in Behavioral Cloning.
Dear Ladies and Gentlemen,
We are releasing a new survey on Deep Generative Models for Robotics 🐣
We tackle timely challenges for robot learning such as what generative models should you use or how to improve the generalization capabilities of BC.
arxiv.org/pdf/2408.04380v3
(3/7) At 13:45 I will talk about Mobile Manipulation Skills & Why We Need Them at the icra-manipulation-skill.gith… and will also be at the panel at 15:30.