Navigation and localization for stair-climbing robot in narrow space

• Main Author: Juspi, Bryant
• Other Authors: Chen I-Ming
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Mechanical engineering::Mechatronics; Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: https://hdl.handle.net/10356/158947
• Institution: Nanyang Technological University

SLAM, formally known as Simultaneous localization and mapping has been gaining a lot of attention recently because it is now feasible to traverse in an unknown environment without any human operator. The improvements in exploring an unknown environment, responding to unforeseen variables and the ability to run autonomously give autonomous mobile robot great advantages over conventional industrial robot. However, the prevailing issues with SLAM are the limited amount of time in real world applications and the degree of accuracy. The aim of this project is to research and design an indoor navigation SLAM system which allows robot to localize and navigate around stairs. This project also aims to study the suitable object detection model in use with the SLAM system. The object detection model needs to have a balance between inference speed and accuracy of prediction. Therefore, further studies are needed to optimize the object detection model for the SLAM system.