Ruihan (Multy) Xu 徐睿涵

Graduate student at MIT
Robotics, Computer Sciense, AeroAstro.

prof_pic.jpg

Band practice at SMTD,

University of Michigan,

Ann Arbor, MI 48105

I am a second year graduate student in Laboratory for Information & Decision Systems (LIDS) and Department of Aeronautics and Astronautics (AeroAstro) at MIT advised by Prof. Luca Carlone.

My research interests lie in spatial AI and perception for robots to better understand its surrounding and make informed decision. I’m especially interested in methods for building rich spatial representations for robot autonomy, such as scen graphs, and aiming to bring working robots into complex everyday environments.

Before joingng MIT, I was an undergratue student at the University of Michigan, pursuing degree of Computer Science focusing on Robotics. I feel fortuntate to work in CURLY lab directed by Prof. Maani Ghaffari. We aims to discover methods for computational reasoning and perception that will enable robots to effectively assist people in common human environments. Besides, I also work closely with Prof. Chad Jenkins in The Lab for Progress.

“君子和而不同” – “Scholars seek harmony in diversity”. Therefore, I’m excited to make friends with diversified interest and background, form collabration, and pursue great ideas. Feel free to contact me through email.

news

Aug 22, 2026 Our work Imaginative World Modeling With Scene Graphs Via VLM for Object Navigation has been accepted to RA-L!
Jun 19, 2026 Our work Hydra++: Real-Time Hierarchical 3D Scene Graph Construction With Object-Level Shape Estimation has been accepted to IEEE/RSJ IROS!
Sep 01, 2025 I have officially joined the SPARK Lab at MIT as a graduate student, advised by Prof. Luca Carlone. :tada:
Mar 03, 2025 Our work LatentBKI: Open-Dictionary Continuous Mapping in Visual-Language Latent Spaces With Quantifiable Uncertainty has been accpeted by RA-L.
Oct 03, 2024 Our work LatentBKI: Open-Dictionary Continuous Mapping in Visual-Language Latent Spaces With Quantifiable Uncertainty has been submitted to RA-L for review.