Hai Huang

Harbin Engineering University, Ningbo University

Papers

Total Citations

215

H-Index

About

Hai Huang is a researcher whose work spans robotics, structural optimization, and intelligent transportation systems, demonstrating a remarkable breadth of technical expertise across multiple engineering disciplines. His most influential contribution, the attention-based spatio-temporal graph neural network AST-GNN (2021), has garnered 196 citations and represents a significant advancement in pedestrian trajectory prediction — a critical challenge in autonomous driving and crowd simulation. By integrating attention mechanisms with spatio-temporal graph neural networks, Huang's framework enables more accurate modeling of complex interaction-aware movement patterns among pedestrians, pushing the boundaries of what machine learning can achieve in dynamic real-world environments. Earlier in his career, Huang made contributions to industrial robotics through structural optimization research, notably investigating topology optimization of the L-shape arm of the Motorman-HP20 industrial robot. This work applied multi-body dynamic modeling and finite element analysis to achieve lightweight yet structurally sound robotic components — a practical advancement for manufacturing efficiency and robotic performance. Huang's trajectory reflects an evolution from classical mechanical engineering and robotics toward cutting-edge deep learning applications in autonomous systems, making his work highly relevant to researchers in computer vision, human-robot interaction, and intelligent transportation.