An intent aware framework for semantic edge intelligence in autonomous 6g UAV and vehicular networks

Abstract Autonomous UAV–vehicular communication in 6G environments requires low-latency, context-aware, and scalable edge intelligence beyond conventional data-centric transmission models. This paper proposes SEM-EDGE-X, an intent-aware semantic edge intelligence framework that integrates semantic feature extraction, adaptive edge routing, predictive resource allocation, and multi-agent coordination for dynamic UAV–vehicular networks. The framework incorporates cold-start semantic fusion, uncertainty-aware human-in-the-loop reinforcement, semantic conflict resolution, and trust-driven security mechanisms to improve communication reliability and decision efficiency. SEM-EDGE-X is evaluated using a MATLAB/Python-based simulation environment with SUMO mobility modelling, KITTI and UAVDT datasets, and CNN-based semantic encoding under varying UAV and vehicular densities. Performance is analyzed using latency, semantic accuracy, semantic similarity, energy consumption, throughput, scalability, and SNR robustness metrics. Experimental results demonstrate that SEM-EDGE-X consistently outperforms conventional cloud, MEC, DeepSC, and AI-assisted edge frameworks in highly dynamic 6G communication environments