Qingwen Zhang

KTH Royal Institute of Technology, Hong Kong University of Science and Technology

Papers

Total Citations

H-Index

About

Qingwen Zhang is a robotics researcher advancing how machines perceive and navigate dynamic, unstructured environments. Her work centers on real-time mapping, dynamic scene understanding, and uncertainty-aware perception for legged and autonomous systems. In her highly cited 2024 paper, “DUFOMap: Efficient Dynamic Awareness Mapping” (28 citations), Zhang tackles the fundamental challenge of distinguishing static from dynamic elements in real-world scenes, enabling robots to build reliable maps for localization and planning. Her earlier 2022 work, “Real-Time Neural Dense Elevation Mapping for Urban Terrain With Uncertainty Estimations” (18 citations), introduces a novel framework that reconstructs complex urban terrain using neural representations while quantifying prediction uncertainty—a critical capability for safe legged robot locomotion and navigation. By integrating deep learning with probabilistic estimation, Zhang’s research bridges the gap between robust perception and real-time performance. Her contributions are shaping the next generation of autonomous systems that can operate reliably in crowded, unpredictable environments, making her a rising leader in the field of dynamic awareness and terrain mapping for robotics.