REALM
Welcome to REALM!
Welcome to the Reliable Autonomous Systems Lab at MIT (REALM)! We design, analyze and verify safe control systems. Our work lies at the intersection of control theory, machine learning, and formal methods, with a focus on safety in systems with nonlinear, high-dimensional and difficult-to-model dynamics.
Recent News
- [Jan'26] Seven papers accepted at ICLR 2026
- [Sep '25] Two papers accepted at NeurIPS 2025
- [Jun'25] Oswin and Songyuan won RSS 2025 Outstanding Student Paper Award
- [Apr' 25] Two papers accepted at RSS 2025
- [Jan' 25] Five papers accepted at ICRA 2025
- [Jan' 25] Four papers accepted at ICLR 2025
- [Sep'24] One paper accepted at Neurips
- [Aug'24] Oswin got Matlab Fellowship
- [Jan'24] Charles got the best presenter award for the 29th Annual LIDS Student Conference (Autonomy, Control, & Applications session)
- [Nov' 23] Charles presented at the MIT Energy & Climate Club Seminar
- [Oct' 23] One paper accepted at EMNLP
- [Aug' 23] Yue got Amazon Robotics Fellowship
- [Jun' 23] Mingxin got MathWorks Fellowship
Our Missions
We look at autonomy through two complementary points of view:
- Given an autonomous system, how do we know whether it will accomplish its mission safely?
- How can we use safety considerations to design better autonomous systems?
Currently, we are thinking a lot about how these questions can apply to learning-enabled autonomy but we are also interested in exploring these questions more broadly.
