Graph-SLAM Techniques for Mobile Mapping in Indoor and Urban Environments
Students should have strong programming skills in Python and C++. Knowledge about Linux and docker is useful, but not necessary. The multi-sensor backpack systemhas 2 x VLP16 lidars, Ladybug5+ HD camera, BD990 Trimble GNSS, and XSens MTI 630R IMU.
Simultaneous Localization and Mapping (SLAM) techniques, especially graph-SLAM techniques, can be used for mobile mapping. Mobile mapping means collecting accurate 3D data in indoor and urban environments. However, accurately mapping these environments can be challenging due to complex geometry, dynamic objects, and sensor noise. The goal of this thesis is to review the state of the art in SLAM techniques and develop improved methods for mobile mapping in indoor and urban environments using a multi-sensor system.
The objective of this thesis is to develop advanced SLAM techniques for mobile mapping in indoor and urban environments using a multi-sensor system provided by the laboratory. This will involve a review of the state of the art in SLAM, including recent developments in optimization methods, and sensor fusion. The student will then develop and implement improved SLAM methods in C++ and Python, using datasets provided by the laboratory or collected by the student.
Specifically, the state-of-the-art linear-inertial odometry uses discrete-time techniques to optimize trajectories. Previously, however, it has been shown that continuous-time techniques can provide a more accurate result. Here, the state-of-the-art EKF open-source code, FAST_LIO, [1] will be modified to transfer the original timestamps to the output file. The odometry result with timestamps will then be read with a graphSLAM code developed in this work to optimize the output, which is similar to [2], but leverages the new multi-sensor sensor configuration, the new odometry module, and the lately developed CERES[1] non-linear optimizer or equivalent to output fast and accurate maps. The performance of the new methods will be evaluated and compared against existing SLAM techniques [3]. The techniques developed and implemented by the student will be an integral part of the ITC research efforts focusing on mobile mapping.
[1]: Xu, W., Cai, Y., He, D., Lin, J., & Zhang, F. (2022). Fast-lio2: Fast direct lidar-inertial odometry. IEEE Transactions on Robotics, 38(4), 2053-2073.
[2]: Karam, S., Lehtola, V., & Vosselman, G. (2021). Simple loop closing for continuous 6DOF LIDAR&IMU graph SLAM with planar features for indoor environments. ISPRS journal of photogrammetry and remote sensing, 181, 413-426.
[3]: Zhou, L., Huang, G., Mao, Y., Yu, J., Wang, S., & Kaess, M. (2022). $\mathcal {PLC} $-LiSLAM: LiDAR SLAM With Planes, Lines, and Cylinders. IEEE Robotics and Automation Letters, 7(3), 7163-7170.