Papers
10
Total Citations
334
H-Index
7
About
Dongchun Ren is a leading researcher in intelligent robotics and autonomous systems, with a primary focus on trajectory prediction and mobile robot navigation. His most impactful work centers on developing deep learning architectures that capture complex social interactions in crowded environments. Ren’s landmark paper, “AST-GNN: An attention-based spatio-temporal graph neural network for Interaction-aware pedestrian trajectory prediction,” has garnered 196 citations, establishing a new standard for modeling pedestrian dynamics. He further advanced the field with “Tra2Tra,” a trajectory-to-trajectory prediction framework using global social spatial-temporal attention, and the “CSR” model, which integrates cascade conditional variational autoencoders with socially-aware regression. Ren’s contributions extend to reinforcement learning, where he proposed an OCBA-based method for efficient sample collection, and to classical robotics with a modified artificial potential field algorithm that resolves local minima in path planning. His work on robust trajectory forecasting for multiple intelligent agents and incremental active learning has been widely recognized for addressing real-world challenges in autonomous driving and social robot navigation. With over 330 total citations, Ren’s research continues to shape how autonomous systems understand and predict human behavior in dynamic scenes.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10