An intelligent self-learning drone assistance approach towards V2V communication in smart city

The objective of the study is to investigate the efficient packet transfer among vehicles in the smart city. With the evolution of Intelligent Transportation Systems (ITS), Vehicle to Vehicle (V2V) communication is becoming more prominent for safety and non-safety related applications. The V2V communication facilitates vehicles to interconnect with each other to support numerous applications that will be highly helpful for drivers and passenger's welfare. However, due to the extreme dynamic nature of transportation, the Vehicular Ad hoc Network (VANET) faces many challenges for transferring information among vehicles. Therefore, efficient clustering and dynamic routing are becoming a supreme area of improvement to increase the Packet Delivery Ratio (PDR) and reduce the End-to-End delay for data transfer. In order to overcome the major obstacle, in this paper, we propose an intelligent self-learning approach-based hybrid clustering by integrating Adaptive Network-based Fuzzy Inference System (ANFIS) and dynamic Dijkstra routing for packet transfer between vehicles. Also, experiments were carried out to support the data transfer with the help of drones to provide higher coverage in high dynamic vehicle mobility scenarios. The proposed algorithm is modeled, trained, and tested for performance evaluations metrics such as Packet Delivery Ratio (PDR), End to End delay, CH selection delay.