Hybrid Fuzzy Logic and Ant Colony Optimization-Based Routing for Dynamic VANET Topology Management

VANET facilitates intelligent communication between mobile and stationary vehicles through wireless connections, serving as a crucial component of ITS. A dynamic and self-organising network is established by each vehicle, which operates as a node. Enhanced vehicle safety and passenger comfort are beneficial byproducts; nevertheless, developing efficient dynamic routing protocols is challenging due to the rapid advancement of VANET architecture. This research introduces a novel hybrid model, Fuzzy-ACO-GA, to enhance routing decisions in dynamic vehicular situations. This model incorporates Fuzzy Logic, Ant Colony Optimisation, and Genetic Algorithms. The proposed methodology comprises three phases: data preparation, feature extraction, and model training. Data quality is enhanced during preprocessing by the imputation of missing values and the normalisation of features. A customised fitness function or the PSO approach is employed to determine cluster heads. Fuzzy logic enhances path optimisation and decision-making under uncertainty when utilised with ACO and GA. The experimental findings indicate that the Fuzzy-ACO-GA model attains an impressive 98.54 % accuracy, far exceeding the efficacy of current methodologies. The model's capacity to adjust to frequent topological changes is validated, and its potential to enhance dynamic routing in VANETs is emphasised. This work sets a new benchmark for topology-aware routing systems in VANET.