Roger Girgis

Papers

Total Citations

H-Index

About

Roger Girgis is a researcher advancing the frontier of autonomous systems through cutting-edge work in multi-agent motion prediction and sequential set modeling. His primary research areas center on trajectory forecasting, latent variable models, and transformer architectures for robotic and vehicular control. Girgis’s major contribution lies in developing frameworks that jointly model the complex interplay of social, temporal, and contextual information—a critical challenge for safe human-robot interaction. His most cited work, “Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Prediction” (2021, 38 citations), introduces a novel approach that learns a robust representation of the true joint distribution of multi-agent trajectories, enabling more consistent and socially-aware predictions. This work directly addresses the limitations of prior methods that treat social and temporal factors separately. His follow-up paper, “Autobots: Latent Variable Sequential Set Transformers” (2021), further refines these concepts, emphasizing socially consistent future trajectory modeling. Girgis’s research has significant implications for autonomous driving, drone swarms, and collaborative robotics, providing foundational tools for systems that must anticipate and coordinate with multiple dynamic agents in real time.