Bottino, Verdiana (2024) Innovative Solutions of Integrated Systems for Drone Navigation in Urban Environments. [Tesi di dottorato]

Documento PDF PhDTHESIS_Bottino_Cycle37.pdf Visibile a [TBR] Amministratori dell'archivio Download (2MB) | Richiedi una copia

Abstract

Over the past decade, substantial cost reductions and advancements in technology have driven the widespread adoption of Unmanned Aerial Vehicles evolving them from military tools to versatile, economical solutions for various civilian applications such as infrastructure inspections, 3D mapping, surveillance, damage assessment, precision farming, goods delivery, healthcare, and leisure. With their smaller size, lighter weight, and maneuverability, drones can perform aerial operations that manned aircraft find difficult, resulting in cost savings, environmental benefits, and reduced human risk; also, they can carry multiple sensors simultaneously and undertake complex, long-range tasks efficiently. Recent avionics improvements boosted drone performance significantly; nevertheless, as mission requirements become more stringent, new risks and technical challenges, such as high-integrity navigation and collision avoidance, must be addressed to ensure safe and efficient operations in increasingly complex environments. This work presents innovative system architecture configurations to enhance Unmanned Aircraft Vehicles' performance in terms of reliability, accuracy, and robustness across various applications and environmental conditions. The features of Integrated Navigation were tested for accurate positioning purposes in GNSS-degraded scenarios using both customized and commercial drones and embarking Micro-Electro-Mechanical Sensors. Methodologies aimed at improving autonomous navigation through sunlight polarization techniques were thoroughly explored, while significant progress was made in coupling low-cost sensors with Artificial Neural Network models, also leveraging redundancy to enhance performance. In addition, an innovative micromachined sensor was tested for accurate drone airspeed estimation, marking a key advancement in sensor technology. The performance of satellite corrections provided by Galileo's High Accuracy Service in the Precise Point Positioning localization technique was assessed in post-processing mode using open-source commercial software, showcasing the potential for high-precision navigation. Moreover, extensive research was conducted to improve consumer-grade MEMS sensor performance through advanced sensor fusion and the application of Additive Manufacturing technology, pushing the boundaries of their practical use. A machine learning-based approach was developed to predict flight time and optimize battery consumption in drone-based applications; finally, a preliminary architecture design was proposed to explore the potential of the integration of 5G technology with a GNSS module, laying the foundation for future advancements in drone operations and communications.

Downloads

Actions (login required)

Modifica documento