Tsu-Jan Hsieh

About

Tsu-Jan Hsieh is a researcher focused on advancing autonomous driving systems through trajectory prediction and decision-making under uncertainty. His key research areas include deep learning for motion forecasting, generative adversarial networks (GANs), and safety-critical perception for intelligent vehicles. Hsieh’s most notable contribution is the development of a Social Conditional Generative Adversarial Network (SCGAN) for trajectory prediction at unsignalized intersections—a challenging scenario where vehicles must infer the intentions of others without traffic signals. This work, published in 2021, addresses the limitations of deterministic models by generating probabilistic, socially-aware predictions that capture the inherent uncertainty of real-world driving. With 5 citations to date, his paper has laid groundwork for more robust, proactive driving strategies that prioritize defensive behavior and accident prevention. Hsieh’s research is particularly impactful for students and engineers working on autonomous vehicle safety, as it bridges the gap between theoretical generative models and practical, real-time navigation in complex urban environments. His contributions underscore the importance of modeling social interactions among road users to achieve truly intelligent and safe autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Trajectory Prediction at Unsignalized Intersections using Social Conditional Generative Adversarial Network

   5 citations · 2021

Key Collaborators

Contact & Links