Robotics PhD Student @core_robotics and @GeorgiaTech

Atlanta, Georgia
Joined August 2011
Delighted to cap this year off holding @Seancye and @Pradyumna_T. Their legacy for the @core_robotics is immense, pioneering work on generative AI (diffusion models and LLMs). Your leadership will be missed. @zoox and Toyota Research Institute are lucky to have you! To check…
2
4
17
1,397
Something I’m proud of with the @core_robotics is how collaborative we are. The papers is shared could not have been done without the amazing work of @KinManLee7, Zixuan Wu, Manisha Natarajan, @Cosmo_Xiao, @zulfiqarzaidi95, as well as two excellent, outside collaborators at…
1
1
4
346
Diffusion models can forecast trajectories, which can help with predicting aircraft flight paths to helping robots generate motion plans in complex environments. Check out our latest work on this topic that Dr. Sean Ye presented #IROS2024! Teaser Talk: tiny.cc/nwjrzz
1
4
15
1,093
Congrats to Dr. Sean Ye and Dr. Pradyumna Tambwekar on defending their PhDs on "Trajectory Modeling using Generative Approaches for Scheduling, Planning, and Multi-Agent Systems" and "Human-Centered Specification and Explanation for Mixed-Initiative Interactions", respectively!🎉
1
2
8
331
Sean Ye retweeted
In a little over a week, I'll be sharing 5 years worth of learnings about robot painting, including cable robot control, optimization, and embodied generative AI! (deets 👇) Additionally, if you want to make your very own graffiti art with my robot, DM me within the next week :)
2
2
1
17
1,794
Can Diffusion Models help robots predict agent behavior (a problem in autonomous vehicles (AVs) to narcotics interdiction)? Yes! @Seancye shared his pioneering work @MultiRobotSys in what may revolutionize imitation learning, IRL, LfD, etc. A🧵1/n Paper: arxiv.org/abs/2307.06244
1
5
1
33
3,789
Come play catch with ESTHER if you’re at the @corl_conf tomorrow morning AND afternoon at the Demos! @core_robotics @ICatGT @zulfiqarzaidi95 @MatthewGombolay
5
1
22
3,039
Ever wondered if it's possible to hijack a robot team through communication sabotage? Our #CoRL2023 paper reveals how an adversary communication can coerce robots into unsafe actions by altering their message exchanges, all while learned completely offline in a black box manner.
1
3
6
1,058
We're outsmarting robot teams without peeking at their playbook! By using the surrogate policies to estimate the ground truth reward, we craft an adversary that can scramble communications, all without ever needing to see their actual reward signal.
1
1
1
194
Dive into our research on robot team communication and adversary tactics in our latest paper: openreview.net/pdf?id=bIvIUN… w/Zixuan Wu, Byeolyi Han, and @MatthewGombolay for an in-depth discussion tomorrow (11/7) at our poster and Oral Session 2 (1:45-2:45)! @core_robotics #CoRL2023
1
3
285
Replying to @core_robotics
The @CoreRoboticslab at #IROS2023 had 3 paper presentations, 1 demo, + 1 workshop organized. Kudos to my amazing students for being tremendous ambassadors of our work. Thanks also @gtcomputing for putting together this sweet site highlighting the impact: sites.gatech.edu/research/ir…
1
2
17
3,216