Visual localization and mapping for indoor drone navigation

• Main Author: Goh, Jia Wern
• Other Authors: Mir Feroskhan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Drones; Localization and mapping; Autonomous navigation
• Online Access: https://hdl.handle.net/10356/176524
• Institution: Nanyang Technological University

This paper evaluates the feasibility of developing Visual Localization and Mapping capabilities for Indoor Drone Navigation. The software set-up includes implementing Visual SLAM algorithms which are ORB-SLAM 2 and VINS-Fusion as well as simulation of three state-of-the-art Path Planning algorithms which are Kinodynamic Path Planner, Topological Path Planner and EGO-Planner. Quantitative and comprehensive comparison will be conducted and the most suitable Visual SLAM algorithm and Path Planning algorithm will be selected for physical implementation on the PX4 Vision Autonomy Development Kit Drone. Subsequently, the drone will utilize these algorithms to build a map of the surrounding environment in real time.