Making impossible optimizations possible, at scale and in practice.

TILOS attended the Summit for AI Institutes Leadership (SAIL) conference October 6-9, 2024 in Pittsburgh, PA. On October 10 we showcased TILOS research and K-12 outreach activities at AI Institutes Expo Day, an opportunity for the public to learn about all 25 (and counting) national AI research institutes.

The exposition hall at AI Institutes Expo Day, October 10 at Carnegie Mellon University in Pittsburgh, Pennsylvania, featuring all 25 national AI research institutes.

Networks team co-lead Tara Javidi (center) represented TILOS on the panel Overcoming AI Development Challenges: From Research to Deployment.

TILOS Outreach Coordinator Saura Naderi presents a K-12 outreach project: a robotic flower that "blooms" when you smile at it (and wilts when you stop).

Dr. Yulun Tian is a TILOS postdoctoral scholar at UC San Diego, where he works with TILOS Robotics team member Nikolay Atanasov.

Where did you earn your graduate degree?

My PhD is in AeroAstro from the Massachusetts Institute of Technology.

What are your research interests?

I am passionate about developing scalable and trustworthy robotic systems that operate robustly over long periods of time without human intervention. My current research applies tools from nonlinear and distributed optimization, machine learning, and graph theory to develop principled algorithms with theoretical guarantees and real-world systems for robot perception and navigation.

What type of career will you pursue after your current position?

I will be continuing in academia, joining the Robotics Department at the University of Michigan as an Assistant Professor in Fall 2025.

What type of career will you pursue after your current position?

I won the 2022 IEEE Transactions on Robotics (T-RO) King-Sun Fu Memorial Best Paper Award for my research on multi-agent metric-semantic SLAM. In the paper, we develop a system called Kimera-Multi that is one of the first to enable multi-robot metric-semantic mapping. The system is not only able to map the geometry of the environment such as where obstacles are, but also annotate the map with semantic labels such as "road", "building", etc., helping the robot to achieve a higher level of 3D scene understanding.

What do you enjoy doing when not working on research?

I enjoy outdoor activities such as hiking and playing tennis with friends.

Recordings of all TILOS Seminars and presentations are at tilos.ai/recorded-talks.

TILOS Foundations team member Misha Belkin of UC San Diego will be panelist at this workshop.

Co-organized by TILOS Foundations team member Stefanie Jegelka, associate professor at MIT.

Dr. Wang is an Associate Professor of Data Science and affiliate faculty in the department of Computer Science and Engineering at UC San Diego.

His research interests include statistical theory and methodology, (nonconvex, online) optimization, differential privacy, reinforcement learning, and (theory of) deep learning.

Dr. Yu is an Associate Professor of Computer Science and Engineering at UC San Diego. Her research interests lie primarily in deep learning, especially for large-scale spatiotemporal data.

Rose is broadly interested in high-dimensional optimization, uncertainty quantification and decision making in spatiotemporal environments.

Dr. Drusvyatskiy's research focuses on designing and analyzing algorithms for large-scale optimization problems arising in data scientific domains.

Particular topics of interest include nonconvex optimization in deep learning, stochastic optimization under distributional shifts, and dynamic decisions under uncertainty.

Dr. Hsu is an Associate professor of Computer Science at Columbia University.

He works on algorithmic statistics and machine learning/AI, with the goals of designing efficient algorithms for learning and data analysis, and understanding the limits of efficient computation for these tasks.

Penn Engineering research discovers critical vulnerabilities in AI-enabled robots to increase safety and security

Work led by recent Penn Engineering Ph.D. graduate Alexander Robey, with TILOS Robotics team members Hamed Hassani and Vijay Kumar, finds that AI-governed robots can easily be hacked. READ THE PRESS RELEASE

An interview with TILOS Robotics team member Hao Su

Dr. Su is an Associate Professor of Computer Science and Engineering at UC San Diego, and cofounder and chief technology officer of Hillbot AI, a start-up developing cutting-edge embodied AI technology. Read the article in AsiaTechDaily.

TILOS Robotics team members present new research at IROS 2024, October 14-18 in Abu Dhabi

OSM vs HD Maps: Map Representations for Trajectory Prediction
Jing-Yan Liao, Parth Doshi, Zihan Zhang, David Paz, Henrik Christensen

Harmonic Mobile Manipulation
Ruihan Yang, Xiaolong Wang, et al.

Enhancing Online Road Network Perception and Reasoning with Standard Definition Maps
Hengyuan Zhang, Henrik Christensen, et al.

A Deep Signed Directional Distance Function for Object Shape Representation
Ehsan Zobeidi and Nikolay Atanasov

Dr. Irene Tsapara joins the TILOS Education and Workforce Development team

Dr. Tsapara is the Academic Director of the Data Science Ph.D. program at TILOS member institution National University, and adjunct faculty at Northwestern University School of Professional Studies.

TILOS Foundations team member Fan Chung Graham recipient of the 2024 Revelle Medal

Dr. Graham is a Distinguished Professor Emerita of Mathematics and Computer Science & Engineering at UC San Diego. The Revelle Medal is a prestigious award recognizing sustained, distinguished and extraordinary service to campus. 

TILOS Robotics team co-lead Henrik Christensen interviewed for the Robot Report podcast

Dr. Christensen discusses the intersection between shifting federal policies and the robotics industry and emphasizes the need for more attention and funding for the sector on the October 11, 2024 episode of the Robot Report podcast.

LISTEN TO THE EPISODE

Open-source Semiconductor Chip Design Tool Celebrates Success

The Foundations and Realization of Open, Accessible Design (OpenROAD) project, lead by TILOS Chips team member and distinguished professor at UC San Diego Andrew B. Kahng, democratizes silicon innovation for both students and chip designers. Read the article in UC San Diego Today.

What Is Embodied Intelligence and What Can It Do

Researchers discussed the link between artificial intelligence and robotics at the San Diego Robotics Forum, held at the UC San Diego campus on September 19, 2024. Read the article in UC San Diego Today.

MORE NEWS >

TILOS research publications are available on Google Scholar.

Stay connected with TILOS