Scholar
Yixi Cai
Google Scholar ID: lEGLRrUAAAAJ
Postdoctoral Fellow, Division of Robotics, Perception and Learning, KTH
Citations & Impact
Citations
2,093
H-index
14
i10-index
14
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Paper 'Efficient and Distributed Large-Scale Point Cloud Bundle Adjustment via Majorization-Minimization' accepted for publication in International Journal of Robotics Research (IJRR).
- Started serving as Associate Editor for Robotics and Automation Letters (RA-L).
- Paper 'DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method' accepted as Spotlight in NeurIPS 2025.
- Paper 'A Survey on LiDAR-based Autonomous Aerial Vehicles' accepted for publication in Transactions on Mechatronics (T-Mech).
- Paper 'A High-Payload Robotic Hopper Powered by Bidirectional Thrusters' accepted for publication in Transactions on Robotics (T-RO).
- Paper 'Slope inspection under Dense Vegetation Using LiDAR-based Quadrotors' accepted for publication in Nature Communications.
- Paper 'Temporal Overlapping Prediction: A Self-supervised Pre-training Method for LiDAR Moving Object Segmentation' accepted in ICCV 2025.
- Paper 'Efficient Swept Volume-Based Trajectory Generation for Arbitrary-Shaped Ground Robot Navigation' accepted for publication in International Conference on Intelligent Robots and Systems (IROS 2025).
- Paper 'FAST-LIVO2 on Resource-Constrained Platforms: LiDAR-Inertial-Visual Odometry with Efficient Memory and Computation' accepted for publication in Robotics and Automation Letters (RA-L).
- Paper 'GauSS-MI: Gaussian Splatting Shannon Mutual Information for Active 3D Reconstruction' accepted in Robotics: Science and Systems (RSS) 2025.
- Paper 'Safety-assured high-speed navigation for MAVs' accepted for publication in Science Robotics.
Research Experience
- Currently a Digital Future Postdoctoral Fellow in the Division of Robotics, Perception and Learning (RPL) at KTH Royal Institute of Technology, collaborating with Prof. Patric Jensfelt and Dr. Olov Andersson. During his PhD, he developed ikd-Tree, FAST-LIO2, and D-Map, which have been widely used in the LiDAR community.
Education
- Ph.D. in Robotics from the Mechatronics and Robotic Systems (MaRS) Laboratory at the Department of Mechanical Engineering, University of Hong Kong, supervised by Prof. Fu Zhang and co-supervised by Prof. James Lam; B.S. from the School of Automation Science and Electrical Engineering, Beihang University, graduated in 2020.
Background
- Research Interests: LiDAR-based mapping with applications in Robotics. Bio: Passionate about exploring elegant representations of the world, aiming to unlock boundless possibilities in Robotics.
Co-authors