James Alan Preiss

robotics ph.d. candidate

lab | g-scholar | github | linkedin | arxiv


I am a Ph.D. candidate at the USC Robotic Embedded Systems Lab (RESL) in Los Angeles, CA, advised by Prof. Gaurav Sukhatme. I am interested in both theoretical and applied topics that combine the strengths of machine learning and control theory for robotics with complex dynamics. I am seeking postdoc opportunities beginning Summer 2022.

My past projects include a hardness analysis of trajectory optimization for observability, a method to combine deep dynamics models with traditional control, and several topics in reinforcement learning. Currently I focus on developing the theoretical framework of suboptimal covering numbers to quantify "how much" a good control policy must change with respect to dynamics parameters.

I also co-developed and actively maintain the Crazyswarm platform for multi-quadrotor research. Before joining RESL, I was a software developer for 3d scanning tools at Geomagic and the JMP statistics package. I completed a B.S. in Mathematics and B.A. in Photography at The Evergreen State College in Olympia, WA.

This page last updated October 2021.

publications

code

contact

(first initial) (middle initial) (last name) @ usc.edu