Roshan Dhakal

PhD, Computer Science · George Mason University

I completed my PhD in Computer Science in July 2026 at George Mason University, where I was advised by Prof. Gregory J. Stein in the Robotic Anticipatory Intelligence & Learning (RAIL) Lab.

My work centers on anticipatory task and motion planning for long-lived autonomy — robots and embodied AI operating in persistent, real-world environments. Rather than discarding the planners robots already rely on, I give them a capability they lack: understanding how an action taken now will affect a future they cannot yet see. That understanding lets them act in ways that would otherwise be out of reach, so behaviors like decluttering, preserving accessibility, and preparing a workspace emerge from long-horizon reasoning rather than hand-coded rules.

Before my PhD, I earned an M.S. in Computer Science from George Mason University in 2021 and a B.E. in Computer Engineering from Tribhuvan University in Nepal.