Qingxiang Zhang

Papers

Total Citations

H-Index

About

Qingxiang Zhang is a researcher specializing in robotics, autonomous systems, and 3D spatial data processing, with a particular focus on multi-robot collaborative mapping, semantic map matching, and point cloud registration. His work addresses fundamental challenges in enabling multiple robots to accurately localize themselves and construct shared environmental representations by developing robust algorithms for merging locally generated maps. Zhang's most notable contribution, "LCR-SMM," introduced an Expectation Maximization-based semantic map matching framework capable of handling large initial pose discrepancies — a critical limitation of classical approaches like Iterative Closest Point (ICP). This work, alongside his probabilistic registration model for semantic map fusion, demonstrates his commitment to building more reliable and generalizable solutions for real-world multi-robot deployments. His 2023 work on deep transformer-based point cloud registration further reflects his engagement with cutting-edge deep learning architectures, tackling persistent issues such as noise sensitivity and outlier handling in 3D data alignment. With a growing citation record across robotics and computer vision venues, Zhang's research contributes meaningfully to the foundations of collaborative autonomy. His work will be of particular interest to students and researchers exploring SLAM, multi-agent systems, and perception pipelines for autonomous robots.