The growing reliance on wireless networks and the unique attributes of Flying Ad-hoc Networks (FANETs) have propelled their advancement in both academic and industrial domains. With the proliferation of Unmanned Aerial Vehicles (UAVs), FANETs have become indispensable for diverse applications, including traffic monitoring, videography, and a wide range of military and civilian operations. This study introduces an innovative hybrid metaheuristic swarm intelligence approach integrated with a supervised learning framework utilizing a backpropagation neural network. The proposed method focuses on clustering-based, location-aware, and energy-efficient routing in FANETs. Results demonstrate significant improvements in network performance, including enhanced energy conservation, prolonged network lifetime, and reduced end-to-end delay. Furthermore, the approach achieves high throughput, emphasizing its potential as a robust solution for optimizing network efficiency and sustainability. Additionally, a comparative analysis with existing optimization-based methods highlights issues and significant improvements in metrics like energy consumption, packet delivery ratio, and resistance to attacks. The results reveal that the proposed protocols achieve better energy conservation and prolonged network lifetime without compromising security or quality of service. This research establishes a new paradigm for leveraging supervised learning to address critical challenges in FANETs while promoting energy efficiency and reliability.