I'm a recent graduate of MIT, where I did my B.S. in Electrical Engineering and Computer Science ('25). I'm currently on leave from the M. Eng program. I worked with Zhutian Yang, Leslie Kaelbling, Tomas Lozano-Perez, Ge Yang, and Phillip Isola at CSAIL on hierarchical long-horizon manipulation and learning from virtual demonstrations. I'm interested in a wide variety of robot learning: long-horizon mobile manipulation, synthetic data, theory, etc. I'm currently applying for PhDs in robotics. In my free time, I enjoy playing tennis, lifting, and sailing.

Research

Lucid-XR: An Extended-Reality Data Engine for Robotic Manipulation

Yajvan Ravan*, Adam Rashid*, Alan Yu, Kai McClennen, Gio Huh, Kevin Yang, Zhutian Yang, Qinxi Yu, Xiaolong Wang, Phillip Isola⁺, Ge Yang*⁺

Conference on Robot Learning (CoRL), 2025

project page / paper / video

We build an intuitive VR data collection interface in simulation for imitation learning, producing visual policies from synthetic data that generalize to real-world scenes and amplify data collection capacity 20x.

PoPi: Combining Planning and Diffusion for Mobility with Unknown Dynamics

Yajvan Ravan, Zhutian Yang, Tao Chen, Tomás Lozano-Pérez, Leslie Kaelbling

arXiv preprint

project page / arXiv

Planner-Ordered Policy (PoPi) is an approach for long-horizon robot manipulation problems in which the dynamics are partially unknown, for example, pushing and pulling a 5-wheeled office chair.

LucidSim: Learning Visual Parkour from Generated Images

Alan Yu*, Ge Yang*, Ran Choi, Yajvan Ravan, John Leonard, Phillip Isola

Conference on Robot Learning (CoRL), 2024

website / arXiv / code / video / news video / news

We use generative models as a realistic and diverse data source by augmenting classical physics simulators. We demonstrate that robots trained this way can accomplish highly dynamic tasks like parkour without requiring depth.

Real-Time Wildfire Localization on the NASA AMS using Deep Learning

Yajvan Ravan, Aref Malek, Chester Dolph, Nikhil Behari

AIAA SciTech 2026

paper / dataset

Real-time wildfire detection using multi-spectral NASA imagery. Deep learning model achieves 94.68% accuracy and 84.96% IoU for automated localization through clouds and at night.

Small Projects

BeaverNav

Indoor navigation system for MIT's campus enabling users to navigate between rooms with routes from start to end destination. Built using computer vision, graph algorithms, and A* path planning.

GitHub / Demo

BatterBot

Robotic system for playing baseball using trajectory optimization, inverse kinematics, and kinematic trajectory optimization. Demonstrates nonprehensile manipulation for dynamic aerial objects.

GitHub / Video

Neural Network Pruning with Matrix Methods

Reducing redundant parameters in neural networks using linear algebra techniques including PCA, low-rank approximation, and randomized methods to improve inference time and computational efficiency.

GitHub / Report

MIT Script

Project description coming soon.

GitHub

PGIF with Mamba & S3

Project description coming soon.

GitHub

Resume

Education

Massachusetts Institute of Technology (MIT)

M.Eng in Electrical Engineering and Computer Science

(on-leave)

B.Sc. in Electrical Engineering and Computer Science

Sep 2021 - Jun 2025

GPA: 5.0/5.0

View Coursework

Computer Architecture

Reinforcement Learning

Controls

Algorithms

Computer Vision

Reasoning/Inference in AI

Statistical Inference

Machine Learning

Robotic Manipulation

Experience

MIT CS & AI Lab (Robotics Research)

Aug 2023 - Present

UROP in Learning and Intelligent Systems Group
UROP in Isola Lab

NASA

Jun 2023 - Dec 2023

Computer Vision Research Intern

MIT Media Lab

Oct 2022 - May 2023

UROP in Conformable Decoders

• Worked on wearable eye-tracking sensors using micro piezoelectrics with CC2460 microcontroller

Awards

Gold Medal at International Chemistry Olympiad (IChO) (2019)

USA Math Olympiad 2-time Qualifier (2020-2021)