AI-Driven Innovations in Autonomous Drones: A Thorough Review of Security, Navigation, and Multi-Sensor Technologies

Abstract

Deep learning algorithms are playing an increasingly important role in autonomous drone systems, revolutionizing fields such as agriculture, surveillance, industrial inspection, and security operations. Although these technologies show remarkable potential, they face major challenges in real-time processing, security preservation, and autonomous decision-making. The study aimed primarily to perform a comprehensive systematic review of deep learning applications in autonomous drone systems, focusing on security, navigation, and multisensor analytics. Following the PRISMA methodology for systematic reviews, articles were carefully selected from major scientific databases according to specific inclusion and exclusion criteria. The selected studies were reviewed to identify the different deep learning architectures implemented and the various types of data handled in UAV applications. A total of 17 articles, published between 2021 and 2024, were analyzed following PRISMA guidelines to explore the current state of AI-driven drone technology, examining prevalent methodologies, implementation challenges, and emerging trends. The findings highlight significant advances in processing capabilities, security measures, and domain-specific applications while also identifying critical areas for future research on hybrid architectures, privacy-preserving mechanisms, and efficient multisensor fusion frameworks.