Papers
19
Total Citations
405
H-Index
10
About
Xiaqing Ding is a robotics and autonomous systems researcher whose work sits at the intersection of simultaneous localization and mapping (SLAM), 3D perception, and long-term robot navigation. His research addresses some of the most pressing challenges facing autonomous vehicles and mobile robots, including robust global localization, dense terrain mapping, and multi-sensor calibration in complex, real-world environments. Ding's most influential contribution, "3D LiDAR-Based Global Localization Using Siamese Neural Network" (2019, 105 citations), demonstrated a powerful deep learning approach to localizing autonomous vehicles in 3D point cloud maps without prior pose knowledge — a breakthrough for handling lost-localization scenarios. His work on the GEM elevation mapping system and GPU-accelerated traversability mapping further advanced safe motion planning over unstructured terrain. Notably, his development of the 2-Entity RANSAC framework tackled the persistent challenge of robust visual localization under dramatic environmental changes, while DXQ-Net offered a targetless, differentiable approach to LiDAR-camera extrinsic calibration. Across his publications, Ding has accumulated over 300 citations, reflecting sustained impact across the autonomous robotics community. His body of work collectively advances the reliability and adaptability of robot navigation systems operating across diverse, long-term real-world conditions.
Research Focus
Key Achievements
Top Papers
- 13D LiDAR-Based Global Localization Using Siamese Neural Network105 citations · 2019
- 2
- 3
- 4
- 5
- 6Efficient 3D LIDAR based loop closing using deep neural network24 citations · 2017
- 7GPU accelerated real-time traversability mapping24 citations · 2019
- 8
- 9
- 102-Entity RANSAC for robust visual localization in changing environment11 citations · 2019