FLQL-VANET: a Hybrid of Fuzzy Logic and Q-learning Schemes for QoS Aware Routing in VANET

Vehicular Ad-Hoc Networks (VANETs) play a crucial role in the development of Intelligent Transport Systems (ITS), leveraging Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), and Vehicle-to-Everything (V2X) communication models.These networks, however, face challenges such as node administration, security, and efficient routing, particularly due to the unpredictable nature of vehicle mobility.Traditional routing protocols like Adhoc On-Demand Distance Vector (AODV), Optimized Link State Routing (OLSR), and Destination Sequence Distance Vector (DSDV) are often inadequate for VANETs.Addressing these challenges, we propose FLQL-VANET, a hybrid mechanism that integrates fuzzy logic and Q-learning for enhanced routing.The clustering mechanism in FLQL-VANET selects cluster heads based on parameters such as connectivity, vehicle direction, distance, velocity, path loss, signal quality, and time to leave.Q-learning optimizes the routing path through a reward mechanism, ensuring efficient data transmission.The comparative analysis demonstrates that FLQL-VANET outperforms existing methods such as RMRPTS, ICA-RBF, and KMRP.Our approach achieves an average packet delivery rate of 96.7%, an end-to-end delay of 0.61 seconds, and an average throughput of 868 kbps.These results indicate significant improvements in routing efficiency and reliability, highlighting the potential of FLQL-VANET to enhance VANET performance in real-world scenarios.