Scalable UAV Routing: Integrating Neural Networks with Facility Location and Optimal Path Planning

In UAVnetworks, dynamic environments arise from drones joining and leaving the system, necessitating resilience across various scenarios. This calls for scalable solutions to manage larger networks effectively. Neural Networks provide the adaptability needed to handle these changing conditions, offering a robust approach to maintain stability and performance in UAV networks. We employ the FLPO algorithm to create a dataset for various scenarios and train a neural network, facilitating a model that can address online route prediction after changes in the environment. Currently, we are developing a model that can address the challenges we have on applying available learning algorithms on our dataset.