The paper explores and classifies autonomous unmanned aerial vehicles (UAVs) control systems, focusing on their application in various domains such as military, healthcare, agriculture, and smart cities. It reviews the key autonomy levels defined by SAE and ALFUS frameworks and provides a detailed analysis of both hardware and software architectures, including innovations like microservice approaches and advanced flight control systems. The paper also discusses the use of adaptive control algorithms, machine learning models, and neural networks to enhance UAV performance in dynamic environments. The study highlights the importance of robust cybersecurity measures in UAV systems, addressing vulnerabilities inherent in traditional monolithic architectures. In addition, the research identifies critical challenges in integrating UAVs into existing systems and optimizing them for autonomous missions. Special attention is paid to the potential for autonomous UAVs to perform complex tasks without human intervention, especially in critical scenarios. This paper provides valuable insights for researchers and practitioners looking to design scalable, reliable, and efficient UAV control systems, and can also serve as a guide for future advancement in the field.