Development of simulation platforms and computer vision techniques for vision-based UAV operations

• Main Author: Tan, Tiong Kai
• Other Authors: Low Kin Huat
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Simulation and modeling; Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/167472
• Institution: Nanyang Technological University

The use of Unmanned Aerial Vehicles (UAVs) in today’s modern society has been growing. From surveillance to commercial and personal usage, many types of UAVs have been developed to suit the needs of different parties. To aid development and testing of UAVs, a robust simulation environment is required. The simulation must allow for a realistic visual representation of the actual flight and data collection capabilities for analysis. By developing a realistic and robust environment, the behaviour of UAVs can be carefully studied to minimize operational risks and safety issues that could occur during actual flights. This research project first seeks to develop a simulation software capable of importing simulation environments and conducting test flights through mission planning or manual flight control. After developing the simulation environment, the project then researched computer vision techniques for object detection and classification that would aid development of robust autonomous flight algorithms.