GRASP's Kumar Lab has studied the capabilities of the vision transformer (ViT) architecture that maps depth images to velocity commands in both real-world applications and simulations. Read the study here: www.anishbhattacharya.com/research/vit... #GRASP #GRASPLab #KumarLab

GRASP's Perception, Action, and Learning group has developed "Planning Exploratory Goals" (PEG), a reinforcement learning agent that sets its goals at training time to maximize exploration. Website: penn-pal-lab.github.io/peg/ #GRASP #GRASPLab #PAL

GRASP's Kumar Lab developed the first static-obstacle avoidance method for quadrotors using onboard monocular event cameras. Website: www.anishbhattacharya.com/research/evfly Video: www.youtube.com/watch?v=NI51... #GRASP #GRASPLab #KumarLab

GRASP's own Dr. Eric Eaton (@ericeaton.bsky.social)will serve as managing director for the NSF AI Institute for Human-AI Cooperation (NSF HAIC @haicinstitute.bsky.social ), a national hub for human-centered AI! Read Here: www.grasp.upenn.edu/news/nsfhaic/ #GRASP #GRASPLab #NSF #HAIC

I’m honored to serve as Managing Director of the new NSF AI Institute for Human-AI Cooperation (HAIC), developing adaptive, collaborative, and trustworthy AI systems designed to strengthen human decision-making. @haicinstitute.bsky.social penntoday.upenn.edu/news/penn-pa...

Penn partners in new National Science Foundation AI Institute for Human-AI Cooperation | Penn Today

Funded by a $21.5 million, five-year award, NSF HAIC will focus on pioneering AI’s role in improving healthcare.

penntoday.upenn.edu

Articulate-Anything is a system that automates the articulation of objects from different inputs that include text, image, and videos into high-quality simulations used to train robotic skills. Developed by GRASP's Perception, Action, and Learning Group. Website: articulate-anything.github.io #GRASP

GRASP's Figueroa Lab developed Elastic-DS, a novel Dynamical System that can adapt to new task instances and trajectories. Website: sites.google.com/view/elastic... #GRASP #GRASPLab #FigueroaLab

We hope the Laboratory for Research on the Structure of Matter (LRSM) and Summer Mentorship Program (SMP) had a wonderful time touring the GRASP Lab! A HUGE Thank you to our wonderful GRASP volunteers who showcased their work to these bright students! #GRASP #GRASPLab #SMP #LRSM

Scaffolder is a system that enhances reinforcement learning training by exploiting privileged sensing and outperforms or compares to already existing baseline policies. This policy was created by GRASP's Perception, Action, and Learning Group. Website: penn-pal-lab.github.io/scaffolder/ #GRASPLab

GRASP's Kumar Lab has developed a robotic reinforcement learning system that reduces the Markov decision process and allows for accelerated training. Website: pratikkunapuli.github.io/EQTrackingCo... Video: www.youtube.com/watch?v=AosG... #GRASP #GRASPLab #KumarLab

We want to thank the NATO Allied Command Transformation (ACT) delegation for visiting the GRASP Lab! It was a pleasure to showcase our variety of robotics research projects! Read about it here: www.grasp.upenn.edu/news/nato-ex... #GRASP #GRASPLab #GRASPatPERCH

GRASP's Sung Robotics lab developed a lightweight and compact tunable-stiffness actuator based on a coiled spring. The actuator could be useful in future soft robotics applications! Video: www.youtube.com/watch?v=52WV... #GRASP #GRASPLab #SungRobotics

GRASP's Kumar Lab has developed HALO, a system designed to map and explore large-scale outdoor environments with real-time dense 3D reconstruction vision at large distances. Website: tyuezhan.github.io/halo/ #GRASP #GRASPLab #KumarLab

Asymmetric Advantage Weighted Regression (AAWR) is a type of reinforcement learning that leverages privileged information to train active perception policies on real robots. Created by GRASP's Perception, Action, and Learning Group. Website: penn-pal-lab.github.io/aawr/ #GRASP #GRASPLab #PAL

GRASP's own Perception, Action, and Learning Group has developed RoboArena, a community-run benchmark for evaluating generalist robot policies through the DROID robot platform. Website: robo-arena.github.io #GRASP #GRASPLab #KumarLab

Magnetic Origami Reprogramming and Folding System (MORF) is a joint project between GRASP's Sung Robotics and Figueroa Labs consisting of thermoplastic and magnetic laminate sheets that self-fold with an external magnetic field or heating. Video: shorturl.at/ETDcs #GRASP #GRASPLab #SungRobotics

GRASP's Kumar Lab developed EV-Catcher, a system that uses Event-based sensors to estimate impact locations of high-speed objects accurately. Publication: ieeexplore.ieee.org/abstract/doc... #GRASP #GRASPLab #KumarLab

Elastic Motion Policy (EMP) is a framework created by GRASP's Figueroa Lab that allows robots to adapt their movements while continuing their task. Website: elastic-motion-policy.github.io/EMP/ #GRASP #GRASPLab #FigueroaLab

CONGRATULATIONS to GRASP's own Dr. Mark Yim on receiving the IEEE Transactions on Robotics Distinguished Service Award!!! #GRASP #GRASPLab #IEEEAward #ICRA2026

It was fun to show these students our lab!

GRASP Lab@grasplab.bsky.social · last mo.

We hope the students from the Glassman Penn Scholars and Penn Engineering High School Internship programs had a wonderful time touring the GRASP Lab and seeing the amazing work from our students! #GRASP #GRASPLab #GRASPTour #HIP #GPS

We hope the students from the Glassman Penn Scholars and Penn Engineering High School Internship programs had a wonderful time touring the GRASP Lab and seeing the amazing work from our students! #GRASP #GRASPLab #GRASPTour #HIP #GPS

GRASP's Sung Robotics and ScalAR Lab have been working on SALP, an underwater robot inspired by ocean invertebrates of the same name. Paper Link: iopscience.iop.org/article/10.1... Video: www.youtube.com/watch?v=0D7Z... #GRASP #GRASPLab #SungRobotics

We hope everyone had a wonderful time at the GRASP Reunion at ICRA 2026 in Vienna! Hope to see you all again at the next GRASP Reunion!!! #GRASP #GRASPLab #GRASPReunion #ICRA2026

GRASP's Figueroa Lab is developing the VPP-TC Viability-Preserving Passive Torque Control. This framework ensures the robot remains safe and can avoid collisions effectively. Website: vpp-tc.github.io/webpage/ Video: www.youtube.com/watch?v=GWP8... #GRASP #GRASPLab #FigueroaLab

RoboGuard is a general-purpose guardrail for large language model-enabled robots being developed by GRASP's Kumar Lab. Website: robo-guard.github.io Video: www.youtube.com/watch?v=uhd6... #GRASP #GRASPLab #KumarLab

VLMgineer is a framework that combines vision-language models with evolutionary search that outperforms existing human-crafted tools. Project of GRASP's Figueroa Lab and the Perception, Action, and Learning group. Link: vlmgineer.github.io/release #GRASP #GRASPLab #PAL #FigueroaLab

Bones and tissue can heal, so why not robots? GRASP's Miskin Lab, a nanorobotics lab, is researching a material that can self-heal when stimulated by electricity. Read more: advanced.onlinelibrary.wiley.com/doi/pdfdirec... #GRASP #GRASPLab #Nanorobotics

GRASP's Kumar Lab created RT-Guide, a framework that actively maps areas using a Gaussian splatting method for real-time navigation. Link: tyuezhan.github.io/RT_GuIDE/ Video: www.youtube.com/watch?v=0Wng... #GRASP #GRASPLab #KumarLab

GRASP's Perception, Action, and Learning (PennPAL) group is working on Eureka, an LLM-powered algorithm that acquires complex skills through reinforcement learning! Link: eureka-research.github.io #GRASP #GRASPLab #PAL

Check out Miskin Lab's research article on "Microscopic robots that sense, think, act, and compute", led by GRASP faculty member Prof. Marc Miskin! Article Link: www.science.org/doi/pdf/10.1... #GRASP #GRASPLab