Abstract
This paper presents a semantic-aware multi-modal resource allocation (SAMRA)for multi-task using multi-agent reinforcement learning (MARL), termedSAMRAMARL, utilizing in platoon systems where cellular vehicle-to-everything(C-V2X) communication is employed. The proposed approach leverages the semanticinformation to optimize the allocation of communication resources. Byintegrating a distributed multi-agent reinforcement learning (MARL) algorithm,SAMRAMARL enables autonomous decision-making for each vehicle, channelassignment optimization, power allocation, and semantic symbol length based onthe contextual importance of the transmitted information. Thissemantic-awareness ensures that both vehicle-to-vehicle (V2V) andvehicle-to-infrastructure (V2I) communications prioritize data that is criticalfor maintaining safe and efficient platoon operations. The framework alsointroduces a tailored quality of experience (QoE) metric for semanticcommunication, aiming to maximize QoE in V2V links while improving the successrate of semantic information transmission (SRS). Extensive simulations hasdemonstrated that SAMRAMARL outperforms existing methods, achieving significantgains in QoE and communication efficiency in C-V2X platooning scenarios.