Yuehai Chen

Papers

Total Citations

H-Index

About

Yuehai Chen is a leading researcher in autonomous navigation and pedestrian trajectory prediction, with a focus on developing AI-driven models that enable robots and self-driving vehicles to safely navigate crowded environments. His work addresses the critical challenge of anticipating human movement in complex, dynamic scenes, where subtle social interactions and unpredictable behaviors pose significant risks. Chen’s most influential contribution is the **Interaction-Aware LSTM (IA-LSTM)** model (2024, 43 citations), which integrates social cues into long short-term memory networks to predict pedestrian paths with high accuracy, directly informing collision-avoidance policies. He also introduced **Tra2Tra** (2021, 36 citations), a global social spatial-temporal attentive neural network that captures both local and crowd-level interaction patterns, setting a new benchmark for robust, real-time trajectory forecasting. Through **incremental active learning** (2021), Chen further advanced the field by enabling models to adaptively learn from sparse, ambiguous data, reducing annotation costs while maintaining prediction reliability. His research bridges deep learning and robotics, with direct applications in autonomous driving, service robots, and smart surveillance. With a growing citation impact, Chen’s work is foundational for creating safer, more socially aware autonomous systems.