Answers are grounded in this paper and cite their sources.
Vehicular Ad Hoc Networks (VANETs) are an important component of Intelligent Transportation Systems (ITS), enabling real-time communication to support road safety, mobility, and transportation efficiency. However, dynamic network topology, stringent latency requirements, security and privacy threats, communication overhead, and resource constraints continue to challenge reliable vehicular communication. Artificial Intelligence (AI) provides data-driven approaches for intelligent routing, adaptive communication, intrusion detection, predictive maintenance, and distributed decision-making. This paper presents a systematic literature review of AI applications in VANETs covering studies published from January 2018 to 30 June 2026. Following the PRISMA 2020 methodology, the literature is organized into four themes: AI-based routing and communication optimization, security and privacy enhancement, predictive maintenance and vehicle health monitoring, and emerging AI paradigms for intelligent vehicular networking. The review compares AI techniques, datasets, communication technologies, evaluation environments, performance metrics, and reported limitations. The synthesis highlights the growing adoption of Reinforcement Learning, Deep Learning, Federated Learning, Graph Neural Networks, and Edge Intelligence across these domains. Key challenges include data heterogeneity, dynamic environments, scalability, communication overhead, privacy, explainability, resource constraints, and reproducibility. Future directions include federated edge intelligence, dynamic graph learning, AI-native 6G, semantic communication, Digital Twins, multimodal AI, and trustworthy AI for scalable and secure vehicular networks.
No takes yet. Share an insight, caveat, or question.
Mohan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: