Implementing slam for practical scenarios

• Main Author: Lim, Zhi Jian
• Format: Final Year Project / Dissertation / Thesis
• Published: 2022
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://eprints.utar.edu.my/5395/1/MH_1701932_Final_%2D_ZHI_JIAN_LIM.pdf; http://eprints.utar.edu.my/5395/

Mobile robots, especially service robots nowadays are equipped with LiDAR or RGBD cameras for 2D Simultaneous Localisation and Mapping (SLAM) and navigation purposes. However, the state-of-the-art 2D SLAM packages that are available in Robot Operating System (ROS) are prone to environmental factors, such as the presence of noise, the presence of repetitive structures, and the lack of features in an environment. The types of sensors used for mapping would affect the scan matching and loop closure abilities of the 2D SLAM packages. Therefore, this project aims to provide an in-depth understanding of the capabilities, performances, and limitations of the 2D SLAM packages, so that more insights could be provided for successful SLAM implementations. In this project, mapping procedures will be carried out on a service robot in the different scenarios of venues, sensors, and 2D SLAM packages, and the results will be compared for further evaluation. From the quality of mapping, this project would provide insights into the choice of SLAM package, the tuning of SLAM parameters, and the choice of different sensors, based on the nature of the surroundings, to obtain the best configuration that results in the best mapping quality.