Deep Neural Network and Zero Trust Principles for Intelligent Drone Management

As more people become interested in machine learning applications and solutions, crucial decisions involving systems engineering must start considering risks that have not previously been identified. These risks come from potential risks associated with collecting machine learning (ML) data, training, and ML-based conclusions. Even cyber-physical systems like drones and automated factories, automated vehicles, and financial investment systems have benefited from this form of risk analysis. This type of risk analysis has benefited the development of virtual assistants. Consequently, the work aims to investigate and address the potential risks of increasingly automated systems by using sophisticated ML techniques in systems engineering. The research, therefore, provides a zero-trust model with a deep neural network (also known as ZTM-DNN) in a smart agricultural system. In this technique, zero-trust concepts are applied to the management of drones in real-time to achieve smart farming. The suggested model is tested via simulation to determine whether it should be recommended in the real-time, physical agricultural system. The simulation results show that the suggested model compared to the existing methods, increased efficiency in security management by <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 8. 5 6 \%}$</tex>.