I study how to make autonomous perception pipelines smaller, faster, and more robust — from BEV distillation and multimodal fusion to RAW-to-task sensing co-design and VLA policies under sensor degradation. As a Postdoctoral Scholar at UCF, I am extending this work into AI-powered urban mobility, spatiotemporal trajectory modeling, and generative digital-twin simulation.
From edge-deployable perception to city-scale mobility models — always deployment-first, task-driven, and experimentally grounded.
