Work

I was one of the developers of XIVO, a visual-inertial odometry system developed at UCLA. and the Temporal Logic Planning toolbox at Caltech.

The rest of my open-source code is on Github.

S. Tsuei, S. Soatto, "Quantifying VIO Uncertainty", Mar. 2023. preprint available.
S. Tsuei, W. Mo, S. Soatto, "Feature Tracks are not Zero-Mean Gaussian", Mar. 2023. preprint available.
A. Schperberg, S. Tsuei, S. Soatto, and D. Hong, "SABER: Data-Driven Motion Planner for Autonomously Navigating Heterogeneous Robots," in Robotics and Automation Letters (RAL), Aug. 2021. PDF
S. Tsuei, S. Soatto, P. Tabuada, and M. B. Milam, "Learned Uncertainty Calibration for Visual Inertial Localization," in International Conference on Robotics and Automation (ICRA), 2021. PDF
A. Schperberg, K. Chen, S. Tsuei, M. Jewett, J. Hooks, S. Soatto, A. Mehta, and D. Hong, "Risk-Averse MPC via Visual-Inertial Input and Recurrent Networks for Online Collision Avoidance," in Intelligent Robots and Systems (IROS), 2020. PDF
A. Wong, X. Fei, S. Tsuei, and S. Soatto, "Unsupervised Depth Completion From Visual Inertial Odometry," in IEEE Robotics and Automation Letters, Apr. 2020. PDF
Y. Shoukry, P. Tabuada, S. Tsuei, M. B. Milam, J. W. Grizzle, and A. D. Ames, "Closed-form controlled invariant sets for pedestrian avoidance," in American Control Conference (ACC), 2017.
S. Tsuei and M. B. Milam, "Trajectory generation for constrained differentially flat systems with time and frequency domain objectives," in IEEE 55th Conference on Decision and Control (CDC), 2016.