Harrison Delecki

Papers

Total Citations

H-Index

About

Harrison Delecki is a researcher advancing the frontier of safe, intelligent robotics through probabilistic machine learning. His work centers on developing robust state estimation and motion planning techniques that enable autonomous agents to reason about uncertain environments and the intentions of other agents. Delecki’s most-cited paper, “Deep Normalizing Flows for State Estimation” (2023), introduces a novel framework that leverages normalizing flows—a powerful class of deep generative models—to capture complex, multimodal probability distributions for trajectory prediction. This approach addresses critical limitations of classical estimation methods, offering more accurate and reliable uncertainty quantification for next-generation robotic systems. By enabling agents to better anticipate the movements of others, his research directly supports safer and more efficient autonomous navigation in dynamic, multi-agent settings. With 2 citations to date, this foundational work is already influencing the field. Delecki’s contributions are particularly timely as the demand for trustworthy AI in robotics grows, positioning him as a promising voice in the development of perception and planning algorithms that prioritize both performance and safety.