Riming Xu

About

Riming Xu is a leading researcher in autonomous navigation and human-robot interaction, with a primary focus on pedestrian trajectory prediction. Their most-cited work, "STI-GAN: Multimodal Pedestrian Trajectory Prediction Using Spatiotemporal Interactions and a Generative Adversarial Network" (2021, 41 citations), addresses a critical challenge for autonomous vehicles and socially interactive robots: safely anticipating the future paths of multiple pedestrians. Xu's key contribution lies in developing a novel framework that captures both spatial and temporal interactions between individuals while accounting for the inherently multimodal nature of human movement—recognizing that pedestrians can choose from many plausible trajectories. By integrating spatiotemporal interaction modeling with a generative adversarial network (GAN), this work enables more robust and realistic predictions than traditional deterministic approaches. This research is foundational for ensuring that autonomous systems can navigate crowded, dynamic environments with greater safety and social awareness. Xu's work continues to influence the development of socially compliant navigation algorithms, bridging the gap between computer vision, robotics, and human behavior modeling.

Research Focus

Key Achievements

Top Papers

1. 1
   STI-GAN: Multimodal Pedestrian Trajectory Prediction Using Spatiotemporal Interactions and a Generative Adversarial Network

   41 citations · 2021

Key Collaborators

Contact & Links