🚨Super excited to be presenting our latest work on pose estimation in space 🚀 at #IROS2023@ieeeiros: “6D Object Pose Estimation from Approximate 3D Models for Orbital Robotics” @DLR_de
Paper: https://t.co/akmgqBPRGN
Code: https://t.co/18uz9X5E4C (stay tuned)
Congratulations to *Dr.* @mwoerhe for exceptional work over the past years, an excellent thesis, and an outstanding defense talk!! I was really lucky and am more than grateful to have had you as the first PhD student in my team 🎉🎉🎉
We achieve state-of-the-art on a satellite pose estimation challenge by @esaACT.
Big shoutout to my amazing colleagues who made this work possible @MaxlDur@ma_sundermeyer@StoiberMa and Rudolph Triebel!
Come to our #IROS2023 poster on Wednesday!
🚨Super excited to be presenting our latest work on pose estimation in space 🚀 at #IROS2023@ieeeiros: “6D Object Pose Estimation from Approximate 3D Models for Orbital Robotics” @DLR_de
Paper: https://t.co/akmgqBPRGN
Code: https://t.co/18uz9X5E4C (stay tuned)
After several months of beta, we are happy to announce the release of Stable-Baselines3 (SB3) v1.0, a set of reliable implementations of reinforcement learning (RL) algorithms in PyTorch =D!
Blog post: https://t.co/IvOqj0twCt
GitHub: https://t.co/SlsjdrL1um
Really happy to see our review (and the first paper of my PhD 🥳) 'Multi-omics integration in biomedical research – A #metabolomics-centric review' out!
Thank you to my amazing co-authors @JanKrumsiek, @KastenmullerLab,@_MatthiasArnold.
https://t.co/vPjBHAikTa
Pretty excited about this collaboration. Obstacle avoidance for robotic manipulators is a hard problem, especially on the basis of vision. Our approach allows a seamless integration of a learned policy while maintaining the responsiveness required for many critical tasks.
New Paper @corl_conf :
Learning Vision-based Reactive Policies for Obstacle Avoidance.
Our goal is to learn the vision-motion relationship for high-DoF robot manipulators.
Project: https://t.co/BseXg9Y8Pa
Paper: https://t.co/cpjwDDDSoK
If you ever wondered what makes different implementations of the same RL algorithm perform very differently on the same task, check out our latest #NeurIPS2020 deep RL workshop paper.
Paper: https://t.co/48GJ1OSw1a
Project: https://t.co/zhxkSxEuzB
Code: https://t.co/ZcbNbm7NJy
Really excited about this!
For more information check out our session at #AAIC20 on "Genome-Metabolome-Phenome: Towards an Integrated Molecular Atlas for Alzheimer's Disease"
3/ Their implementation has never seen the configuration of the challenge before and had to adapt to different poses and deteriorating effects, such as occlusion. They overcame the challenge via a robust computer vision pipeline and clever 3-d printed gripper design.
We just won the Robothon at @MSRM_TUM ! 🎉
Special thanks to our supervisors for supporting us. We had a really challenging week and we are looking forward to the next challenge. #Robothon#Hackathon#Robotics#TUM
We finally received the confirmation that the winners of our student @MSRM_TUM #Robothon will get free entry to compete at @automaticafair in June (for a price pool of 50k €)!
Each year, about 7 million tons of plastics end up in the ocean of which a substantial amount is washed ashore. The challenge this year is to build a robust solution that can autonomously separate waste from gravel and sand to accelerate the cleaning process of coastal areas.
Kicking-off this year's Robothon at the @MSRM_TUM! Teams of students compete to come up with innovative ideas for the challenge "#CoastalCleanUp: Robotics-aided Climate Protection". #Hackathon#Robotics