Dissertation defence (Information and Communication Technology): MSc Qingqing Li
Time
31.5.2024 at 12.00 - 16.00
MSc Qingqing Li defends the dissertation in Information and Communication Technology titled “LiDAR Based Multi-Sensor Fusion for Localization, Mapping, and Tracking” at the University of Turku on 31 May 2024 at 12.00 (University of Turku, Agora XXI, Turku).
Opponent: Associate Professor Evangelos Boukas (Technical University of Denmark, Denmark)
Custos: Professor Tomi Westerlund (University of Turku)
Doctoral Dissertation at UTUPub: https://urn.fi/URN:ISBN:978-951-29-9736-7
***
Summary of the Doctoral Dissertation:
This thesis explores LiDAR-based sensor fusion algorithms to address perception challenges in autonomous systems, focusing primarily on dense mapping and global localization using diverse LiDAR sensors. The research involves integrating novel LiDARs, IMUs, and camera sensors to create a comprehensive dataset essential for developing advanced sensor fusion and general-purpose localization and mapping algorithms. Innovative methodologies for global localization across varied environments are introduced, including a robust multi-modal LiDAR inertial odometry and a dense mapping framework, which enhance mapping precision and situational awareness. The study also integrates solid-state LiDARs with camera-based deep-learning techniques for object tracking, refining mapping accuracy in dynamic environments. These advancements significantly enhance the reliability and efficiency of autonomous systems in real-world scenarios.
Opponent: Associate Professor Evangelos Boukas (Technical University of Denmark, Denmark)
Custos: Professor Tomi Westerlund (University of Turku)
Doctoral Dissertation at UTUPub: https://urn.fi/URN:ISBN:978-951-29-9736-7
***
Summary of the Doctoral Dissertation:
This thesis explores LiDAR-based sensor fusion algorithms to address perception challenges in autonomous systems, focusing primarily on dense mapping and global localization using diverse LiDAR sensors. The research involves integrating novel LiDARs, IMUs, and camera sensors to create a comprehensive dataset essential for developing advanced sensor fusion and general-purpose localization and mapping algorithms. Innovative methodologies for global localization across varied environments are introduced, including a robust multi-modal LiDAR inertial odometry and a dense mapping framework, which enhance mapping precision and situational awareness. The study also integrates solid-state LiDARs with camera-based deep-learning techniques for object tracking, refining mapping accuracy in dynamic environments. These advancements significantly enhance the reliability and efficiency of autonomous systems in real-world scenarios.
University Communications