Z-ENG: Quantitative Comparison of Simultaneous Localization and Mapping (SLAM) Algorithms

Félév: 2024-2025 ősz

Kategória: Nincs megadva

Téma leírása

Simultaneous Localization and Mapping (SLAM) algorithms are often employed in robotics systems where sensor data from the system is used to create a local map automatically while simultaneously determining the spatial position of the robot. SLAM methods often rely on iterative computations, supported by traditional statistical methods (such as Kalman filters and particle filters), to create topological maps and offer scalable solutions for localization in the development of autonomous vehicles, considering kinematic constraints.

In practice, the quality of localization is characterized by quantitative factors such as absolute error in current position measurements, loop closure relative error, offset, and required computational performance.

The student's task is to conduct a literature review on available SLAM algorithms, create a comprehensive qualitative comparison of their advantages and disadvantages. Subsequently, a quantitative comparison should be carried out among the available methods (real-time SLAM: monocular and RGB-D SLAM, stereo SLAM) using real-world autonomous navigation datasets.

For this work, the use of publicly available datasets is recommended, and implementation can leverage available libraries like OpenCV (C++/Python) and other image processing tools.

To solve this task, the student will receive assistance from the employees of the Continental AI Development Center.

If you are interested in the topic, be sure to contact Dávid Sik by email before applying, indicating the selected topic, training level, major and the planned project subject.

Külső partner: Continental Autonomous Mobility Hungary

Maximális létszám: 1 fő