GRASP 2026 Graduate Award Winner Round-Up

_{^{Written by Jillian Mallon}}

_{^{Caption for featured photo: Luyang Hu was presented with an Outstanding Academic Award at the Graduate Awards Ceremony by Robotics Graduate Program Chair M. Ani Hsieh (left) and SEAS Dean Vijay Kumar (right).}}

As another academic year comes to an end, some of the outstanding students who excelled in their coursework or research, or made an overall positive impact on the university, are recognized by the Penn community for their efforts. The GRASP Lab would like to highlight the exceptional awardees whose efforts have impacted the lab positively. 

At the School of Engineering and Applied Science Graduate Awards Ceremony on May 13th, master’s students were presented with Outstanding Academic Awards for excellence in coursework and overall academic performance, and Outstanding Research Award for significant contributions to faculty-supervised research projects. Both master’s and PhD students were given Outstanding Teaching Awards for significant contributions to teaching in one or more courses, and Outstanding Service Awards for significant contributions that enhance the social well-being of other Penn Engineering students. Among these awardees were four master’s students and two PhD students in the GRASP community.

Luyang Hu, Outstanding Academic Award, Robotics Master’s

The Outstanding Academic Award for Robotics Master’s students was presented to Luyang Hu for his exemplary coursework. When he took the MEAM 5200 Introduction to Robotics course taught by Prof. Cynthia Sung, his final project team not only won the competition, but also broke the course’s all-time record for robustly stacking 8 blocks on both sides of the workspace. Luyang also took Prof. Antonio Loquercio’s new course, ESE 6510 Physical Intelligence: Science and Systems, which focuses on training reinforcement learning policies for drone racing and sim-to-real transfer. Despite noting that this was one of the most challenging courses that he completed at Penn due to the breadth of practical knowledge on reinforcement learning, behavior cloning, and hands-on robotics experimentation, Luyang’s team also won the course’s final simulation race. 

“The program offers a wonderful set of courses — a perfect blend of breadth and depth,” Luyang explained. “Since robotics is inherently interdisciplinary, having the freedom to freely explore different directions is very important. I didn’t know that I was drawn to manipulation until I had tried all kinds of directions across the curriculum. On top of that, almost every course pairs lectures with substantial projects, so you get to put abstract knowledge into practice. I’m entering the industry as one of the founding members of Origami Robotics, and I feel confident stepping into that role thanks to the knowledge, projects, and research experiences I built up through coursework at Penn.”

Luyang also completed a ROBO 5990 independent study in Prof. Antonio Loquercio’s Janus Intelligent Robots Lab (JIRL) during which he collaborated on the development of RoSHI, a full-body motion capture system that allows you to record agile movement in the wild. He also worked with Prof. Dinesh Jayaraman on a project that uses VLMs and simulation to iteratively refine physics parameters for arbitrary articulated assets. 

“I am extremely grateful to my advisors, Prof. Antonio Loquercio and Prof. Dinesh Jayaraman. I learned the most through the research projects I did with them,” Luyang shared. “They constantly pushed me to explore, read, and question. I’d also like to thank all the faculty members and student TAs who make this program what it is. The courses are wonderful, and the community is truly amazing.”

Benjamin Aziel, Outstanding Academic Award, MEAM Master’s

From the Mechanical Engineering and Applied Mechanics (MEAM) Master’s program, the Outstanding Academic Award was presented to Benjamin Aziel. At the start of his master’s program, his primary research interest was robot control, which drew him to courses that would help hone that strength. 

“As someone focused on robot control, the courses I took at Penn have given me a fairly comprehensive picture of the state of the field and the tools to reason about it carefully,” Ben explained. “MEAM 5170, Control and Optimization with Application in Robotics with Prof. Michael Posa, in particular, gave me a principled framework for thinking about the full control stack and where different algorithms belong within it. ESE 6500, Learning in Robotics with Prof. Pratik Chaudhari, complemented that well; covering state estimation, reinforcement learning, and imitation learning gave me a much broader sense of the tools available and when each makes sense. Together, they’ve left me feeling well-equipped to reason about algorithm selection, understand the tradeoffs involved, and deploy on hardware. Whether I end up in industry or continue in research, I’m sure that combination of theoretical grounding and practical experience is going to be directly relevant.”

His enthusiasm about his coursework, specifically Prof. Cynthia Sung’s MEAM 5200 Introduction to Robotics class, inspired him to pursue research outside of the boundaries of his initial research interests in Sung Robotics Lab. 

“The breadth of the course gave me a map of the field I could actually navigate from, and the treatment of path planning specifically was what ended up drawing me into Prof. Sung’s lab to work on reactive planning for the NASA TRUSSES project, which I hadn’t anticipated coming in primarily interested in control,” said Ben. “The labs and final project also forced me to think about working with hardware systematically, utilizing the SIL simulation pipeline before running code on the Franka arm. That ended up being incredibly useful when I interfaced with other robot hardware later on, and I ended up TA-ing the course twice after that, once under Prof. Rachel Holladay and once under Prof. Sung.”

Vineet Pasumarti, Outstanding Teaching Award, Robotics Master’s

The Outstanding Teaching Award for Robotics Master’s students was awarded to Vineet Pasumarti, who acted as a teaching assistant for the courses ESE 6500 Learning in Robotics, taught by Prof. Pratik Chaudhari, and ESE 6510 Physical Intelligence, taught by Prof. Antonio Loquercio. Since ESE 6510 was a new course, Vineet played a key role in the inception of this course that is also highly relevant to his personal research interests. 

“As a TA for two large classes, I’m able to interact with an enormous number of students throughout my office hours, recitations, labs, and one-on-one sessions,” Vineet said. “Students in robotics courses, especially at Penn, hail from a diverse set of impressive backgrounds, and it is rewarding to both teach and learn from students of various disciplines.”

The most rewarding part about Vineet’s experience was the final Race Day during the last class of the semester, where students test their new skills in using reinforcement learning to train autonomous racing drones to compete against each other. “Race Day is always an exciting celebration of all the hard work and hoops the students go through in order to succeed. Friendly competition is wonderful to see as a TA,” Vineet remembered.  “It is deeply rewarding to see students who initially struggled with the material ultimately succeed, and even more fulfilling to see students feel confident enough to try challenging and riskier approaches that break out of the mold.”

Vineet will be starting his PhD program in ESE in Fall of 2026, advised by Prof. Loquercio and Prof. Vijay Kumar, to continue his research. “I’m especially grateful to my advisor, Antonio Loquercio, who invited me to take on the TA role for ESE 6510 Physical Intelligence and encouraged me to play a formative role in the course,” Vineet said. “I am also grateful to the other TAs who made teaching the class so rewarding: Chunwei Xing, Jefferson Ng, and Vaibhav Thakkar, whose dedication enabled the students’ success.”

Mateusz Jaszczuk, Outstanding Research Award, MEAM Master’s

The Outstanding Research Award for MEAM Master’s students was awarded to Mateusz Jaszczuk, who completed his thesis research in Prof. Nadia Figueroa’s Figueroa Robotics Lab. 

“I chose to do a thesis because I wanted to experience working on an open-ended problem and explore the full research pipeline — from identifying a limitation in existing impedance controllers, to designing a framework, and implementing it on real robotic hardware,” explained Mateusz. “Other than that, my classes at Penn provided me with skills in optimal control, robot learning and machine learning, all of which were very useful for the project.”

His resulting thesis, advised by Prof. Figueroa and presented to thesis committee members Prof. Rachel Holladay and Prof. Paris Perdikaris, was titled “Learning to Feel: Force-Aware Data-Driven Estimation and Control for Adaptive Physical Interactions.”

“My MEAM master’s thesis focused on improving how torque-controlled robots safely interact with humans in uncertain environments,” explained Mateusz. “Specifically, I developed an adaptive impedance control framework that allows robots to distinguish between payload changes, such as attaching a tool, and external perturbations in real time. The system allows impedance-controlled robots to maintain both safety and high task accuracy.”

Cole Nagata, Outstanding Service Award, MEAM PhD

Outside of the master’s awardees, two PhD Outstanding Service Awards were given to GRASP students this year. From the MEAM department, Cole Nagata received the award for his service on the Executive Board for MEGA, the Mechanical Engineering Graduate Association that coordinates events for MEAM graduate students. The position helped him build community with other students in the department and provide them with a relaxing reprieve from studies.

“An important part of being on the committee was making sure that what we were doing was something approachable, accessible, and enjoyable for everyone, which is something my other executive board members and I made sure we did to the best of our ability,” explained Cole. “Additionally, balancing research with this work was important as well. I always enjoyed seeing people coming to our events and being able to take a break from work and catch up with others, meet new people or just have fun.”

Chris Hsu, Outstanding Service Award, ESE PhD

The Outstanding Service Award for Electrical and Systems Engineering (ESE) PhD students was awarded to Chris Hsu for his 10 semesters of service on the GRASP Students, Faculty, and Industry (SFI) Student Seminar Committee. The committee, established in 2020, is responsible for nominating and inviting external graduate students, robotics industry representatives, and early faculty speakers to speak at our weekly seminar series, planning a day-long itinerary of meetings with GRASP Lab members, and running the seminar. Chris joined the committee in Fall 2022 when there was only one other committee member who co-coordinated 12 seminars per semester. 

“I participated on the SFI committee for most of my time during my PhD,” Chris recalled. “I helped organize the committee during the early years by establishing processes and goals for the seminar. As the years went by I helped to advocate for more members, varying speakers, and involvement by the members of the GRASP community. It has been a rewarding experience being able to participate and hear how the GRASP community has embraced these seminars to interact with rising faculty, industry, and senior student researchers. I got involved in this committee in the first year of my PhD when there was a need for one more student. I figured it would be nice to contribute to the committee and have a say on who to invite. It has grown larger than I could have imagined and it has been a rewarding experience.”

Throughout his time on the committee, Chris not only drafted and updated many of the committee’s foundational documents and procedures, but also hosted a total of 30 speakers! Thanks to his work and advocacy to select student and faculty speakers from a wide variety of disciplines, the committee has expanded to 5 members and has seen increased attendance. 

Bruce Lee, Charles Hallac and Sarah Keil Wolf Award, ESE PhD

Outside of the outstanding student awards, SEAS departments also present annual awards for the most impressive and innovative dissertations of the year. For the ESE department, this award is the Charles Hallac and Sarah Keil Wolf Award, which was presented in 2026 to two recent GRASP Lab alumni: Bruce Lee and Kaustubh Sridhar. 

Bruce Lee, who graduated with his ESE doctorate in 2025, was advised by Prof. Nikolai Matni during his PhD studies. His thesis, titled, “Statistical Limits and Efficient Algorithms for Learning-Enabled Control”, studied the theoretical foundations of reinforcement learning for real-world control systems.

“The main question I studied is: how much real-world experience is needed to learn a good controller?” Bruce recalled. “My work gives mathematical limits and algorithms that help distinguish when learning is feasible from when it is fundamentally data-intensive. In particular, I studied settings where the system only receives partial, noisy observations of the environment, such as from cameras or imperfect sensors, and showed how these limitations can dramatically increase the difficulty of learning.”

Bruce was thrilled that his research was recognized this way after years of development. “The thesis represents many years of effort, but it also reflects the support and influence of many people around me,” Bruce stated. “None of it would have been possible without the mentorship of my advisor, Nik Matni; the outstanding collaborators I have been so privileged to learn from; and the suggestions, encouragement, and friendship of the students in ESE and GRASP.”

In his current position as a postdoctoral scholar at the ETH AI Center, Bruce plans to build on the foundation set by his dissertation. “My current research builds on the insights from my doctoral work to develop scalable and efficient model-based reinforcement learning algorithms,” Bruce explained. “The goal is to use the theoretical understanding developed during my PhD to help establish the algorithmic foundations for the next generation of learning robots.”

Kaustubh Sridhar, Charles Hallac and Sarah Keil Wolf Award, ESE PhD

Kaustubh Sridhar, a 2025 ESE PhD graduate advised by Prof. Insup Lee was also given the Charles Hallac and Sarah Keil Wolf Award for this thesis, “Training Adaptive and Sample-Efficient Autonomous Agents”. Throughout the development of his dissertation, he collaborated with Prof. Dinesh Jayaraman. 

“I am deeply grateful for the unwavering support of my advisor, Insup Lee. I also want to give a special thank you to my long-time collaborators, Dinesh Jayaraman and Souradeep Dutta,” Kaustubh said. “Working closely with them for the majority of my PhD was an extraordinary experience, and they are the reason my research was able to successfully tackle such important problems as in-context adaptation in embodied AI and robotics.”

Kaustubh is continuing his work on building general and adaptive embodied AI agents at his new position as a research scientist at Google DeepMind. “We recently released our work on SIMA 2, an agent that can follow instructions in any 3D environment,” Kaustubh shared. “SIMA 2 can autonomously self-improve in completely new environments. This extends the goals from REGENT, one of the most important papers from my PhD that trained embodied agents to adapt in-context to new environments, to new heights. I plan to continue to bring insights from my PhD to my goal of building general and adaptive embodied AI.”

Jiahui Lei, Morris and Dorothy Rubinoff Award, CIS PhD

One other recent GRASP PhD alumnus was recognized by the School of Engineering for his dissertation this year: Jiahui Lei, who completed his PhD in CIS in 2025 under the advisement of Prof. Kostas Daniilidis. He received the Morris and Dorothy Rubinoff Award, which is presented by SEAS to a graduate degree candidate whose dissertation has resulted in or could lead to innovative applications of computer technology.

Jiahui’s thesis, “4D Vision: Represent, Reconstruct and Generate the Dynamic 3D World”, describes a family of methods and data-driven models to represent, reconstruct and generate the dynamic 3D world and advance the field from 3D computer vision to 4D computer vision.

“I am truly honored to receive this award,” Jiahui shared. “I would like to express my deepest gratitude to my advisor, Professor Kostas Daniilidis, as well as all of my collaborators, for their continuous support, mentorship, and inspiring discussions. The GRASP Laboratory is truly one of a kind. I don’t think there is any other place quite like GRASP, both in terms of its outstanding research atmosphere and its warm, friendly, and vibrant community.”

Jiahui continues to explore this research direction at his postdoctoral position at the University of California, Berkeley. Jiahui said, “I am continuing in exactly the same direction during my postdoc with a specific focus on using the technology from my thesis as a general 4D world model to help robots to plan and act in the real world.”

Alan Zhu, Hugo Otto Wolf Memorial Prize, Physics/CIS Undergraduate, Robotics Accelerated Master’s

Penn Engineering also presents awards to outstanding undergraduate students, and one of these awards was presented this year to a student in GRASP’s Robotics accelerated master’s program, Alan Zhu. Alan was awarded the 2026 Hugo Otto Wolf Memorial Prize, which is awarded to one senior year undergraduate student from each department of the School of Engineering and Applied Science whose research and coursework received the greatest approval of the professors in charge.

“During my senior year I have been working with Professor Liang Feng in MSE and Professor Jonathan Smith in CIS in developing hardware and protocols for the quantum internet,” Alan explained. “As an undergraduate in the Vagelos Integrated Program in Energy Research pursuing physics and computer science degrees, as well as the accelerated ROBO masters, this was the perfect project to unify the background I gained from these experiences to perform a truly novel experiment which we have now published in Science and presented at both the American Physical Society March Meeting and the Internet Engineering Task Force Montreal 124 meeting. The fundamental goal of this project is to generate and distribute quantum entanglement between spatially separated parties with applications in quantum secure communication, time synchronization, and more.”

Throughout his research, he also set up an interconnected system of equipment capable of real-time feedback, which Alan attributes to the knowledge gained through courses and prior research under Prof. Pratik Chaudhari and Prof. Vijay Kumar.

Fengrui Tian, Apple Scholar in AI/ML PhD Fellowship, CIS PhD

Outside of the awards distributed directly from the University of Pennsylvania, one GRASP PhD student received recognition from the robotics industry. Fengrui Tian, a CIS PhD student advised by Prof. Rene Vidal, was one of 20 students internationally nominated by their universities who were accepted to the 2026 Apple Scholars in AI/ML PhD fellowship program. The program supports emerging researchers in AI with funding as they pursue their PhDs, mentorship with Apple researchers, and internship opportunities.

Fengrui is delighted to discover the ways in which the fellowship can help propel his research on 3D and 4D dynamic scene synthesis, with the goal of creating perpetual, consistent world models that can generate immersive environments from limited input.

“I’m incredibly excited that this fellowship includes an upcoming internship at Apple, which will be the cornerstone of my experience. I’m particularly looking forward to collaborating directly with Rick Chang and Oncel Tuzel, whose expertise in 3D/4D vision and generative modeling is directly aligned with my research,” said Fengrui. “Beyond the mentorship, the large-scale GPU resources Apple provides will be transformative for my work. It will allow me to scale my experiments in space generation far beyond what’s possible in an academic setting, moving from individual scenes to high-fidelity, perpetual world models. Ultimately, I hope this collaboration helps bridge the gap between theoretical generative AI and scalable, real-world applications that can perceive and interact with dynamic environments.”

Congratulations again to all of the outstanding achievements of our GRASP Lab members throughout the 2025 to 2026 academic year!