부산대학교 도서관

Multiple LiDAR fusion solutions can significantly enhance a robot's environmental perception, enabling it to navigate through increasingly complex multi-floor architectural structures. Traditional LiDAR-based SLAM approaches often encounter challenges when tasked with mapping across multiple floors, leading to issues such as perception errors, map degradation, and trajectory drift. Therefore, we propose a tightly-coupled multi-LiDAR-inertial SLAM system, leveraging both solid-state LiDARs, along with inertial measurement unit (IMU). By front-end/back-end coupling, the sensor data is optimized, resulting in robust ego-estimation and high-resolution maps in challenging indoor environments. Experiments demonstrate competitive performance compared to state-of-the-art SLAM systems.