Chaoran Zhu
Papers
1
Total Citations
3
H-Index
1
About
Chaoran Zhu is a robotics researcher specializing in 3D perception, place recognition, and deep learning for autonomous systems. Their most notable contribution is the development of CCTNet (Circular Convolutional Transformer Network), a novel architecture that addresses a critical challenge in LiDAR-based place recognition: handling occlusion caused by movable objects. By introducing circular convolution to maintain feature invariance to column-wise shifts in range images—a common issue when dynamic objects block static scene structures—Zhu’s work significantly improves loop closure detection in SLAM and re-localization on prior maps. This innovation bridges the gap between convolutional and transformer-based approaches, offering robust performance in real-world environments with moving pedestrians or vehicles. Though early in their career, Zhu’s research has already garnered attention (3 citations for the 2024 CCTNet paper), reflecting its timely relevance to the robotics community. Their work promises to enhance the reliability of autonomous navigation in cluttered, dynamic settings, making it a valuable reference for students and researchers working on long-term robot autonomy and robust place recognition.
Research Focus
Key Achievements
Top Papers
- 1