RSSI Guided Localization Using Genetic Neural Fusion in Multi-UAV Network

In disaster response scenarios, the formation of multi-Uncrewed Aerial Vehicles (UAVs) networks is a critical task. The dynamic nature of UAVs makes it difficult to employ traditional approaches. This research study presents an improved task allocation mechanism using UAV localization in disaster conditions. Due to the random movement of UAVs localization is quite a complex problem. Therefore, a novel approach is introduced which is based on received signal strength indicator controlled genetic neural fusion. Genetic neural fusion is the combination of a genetic algorithm and an artificial neural network that adjusts weights to improve localization. During simulation 3D environment is properly built for UAVs. The proposed algorithm GenNF is compared with other techniques like CNN-RNN, CNN-LSTM, CNN-GRU, ANN-LSTM, Particle Filter, and Kalman Filter. GenNF has shown better performance regarding the mean square error, root mean square error, and huber loss. RSSI enabled GenNF to accurately calculate signal strength between UAVs and base station. Therefore, if the signal strength is strong then task will be accordingly allocated to UAVs.