Multi-Agent AI Systems for Cooperative Communication in Hybrid MANET-VANET Architectures

Abstract

Hybrid Mobile Ad Hoc Network–Vehicular Ad Hoc Network (MANET-VANET) architectures have emerged as an important solution for enabling reliable, intelligent, and adaptive communication in next-generation wireless networks. The integration of MANET and VANET technologies provides enhanced connectivity, decentralized communication, and efficient data dissemination in highly dynamic transportation environments. However, challenges such as rapid topology changes, routing instability, high mobility, network congestion, packet loss, and security vulnerabilities significantly affect communication performance. To address these issues, Multi-Agent Artificial Intelligence (AI) systems have gained considerable attention due to their capability to provide distributed intelligence, cooperative decision-making, and adaptive communication management. This review paper presents a comprehensive study of Multi-Agent AI systems for cooperative communication in Hybrid MANET-VANET architectures. The paper analyzes the role of intelligent agents in improving routing efficiency, resource allocation, congestion control, trust management, and Quality of Service (QoS). Various AI techniques including Machine Learning (ML), Deep Reinforcement Learning (DRL), Fuzzy Logic, Swarm Intelligence, and Neural Networks are reviewed in the context of cooperative vehicular and mobile ad hoc communication. The study further examines communication models such as Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), and Vehicle-to-Everything (V2X) integrated with intelligent agent-based frameworks. Additionally, the paper discusses existing routing protocols, security mechanisms, and optimization approaches used in Hybrid MANET-VANET systems. Comparative analysis of recent research contributions highlights the advantages of Multi-Agent AI systems in terms of scalability, adaptability, energy efficiency, reduced latency, and improved packet delivery ratio. The review also identifies major research challenges including computational complexity, communication overhead, synchronization issues, and cyber-security threats. Finally, future research directions involving 6G communication, edge intelligence, blockchain integration, and federated learning are explored to support the development of secure and autonomous cooperative communication systems. The findings of this review demonstrate that Multi-Agent AI-enabled Hybrid MANET-VANET architectures offer a promising framework for intelligent transportation systems, smart cities, and next-generation wireless communication networks.