Testing of simulation environments for visual-inertial odometry systems

• Main Author: Ng, Esther Shi Hui
• Other Authors: Wang Han
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/75305
• Institution: Nanyang Technological University

Unmanned Ariel Vehicle (UAV) are commonly used in conditions where sending a human piloted aircraft is undesirable in dangerous environment. In today’s context, UAV are also used as a leisure activity. Flight navigation and control is essential for an UAV. In this project, it aims to provide high accuracy Visual-Inertial Odometry of an UAV through the testing of simulations. Visual-Inertial Odometry determines the position and orientation of an UAV by analysing related camera images. Together with an Inertial Measurement Unit (IMU) that are installed in the UAV, it determines parameters such as linear velocity, angular acceleration, etc. of the UAV. In this Project, Monocular Visual-Inertial Systems (VINS) can also be used to provide state estimation and feedback control for the UAV. Integrating with Linux and Robot Operating System (ROS), packages such as hector_quadrotor and AirSim simulations are used to run the simulations. IMU parameters and camera images will be recorded and be documented for future works.