Abstract
Terahertz (THz) communication is a promising enabler for next-generation wireless networks because it can support ultra-high data rates. However, severe path loss, molecular absorption, and high sensitivity to blockage significantly limit coverage and reliability. To address these challenges, this work proposes a RIS-assisted UAV positioning (RAVP) framework that integrates reconfigurable intelligent surfaces (RIS) with unmanned aerial vehicles (UAVs) and jointly optimizes RIS configuration and UAV deployment to enhance THz communications. RISs provide controllable reflections to improve propagation conditions, while UAVs enable flexible placement of RISs at advantageous locations. A reinforcement learning (RL)-based strategy that combines modified K-means clustering with gradient-based optimization coordinates user grouping, RIS phase-shift adaptation, and UAV positioning within a unified framework. Simulation results show consistent gains in link robustness, achievable data rate, and user connectivity across different network configurations compared with conventional THz systems without RISs or UAV-assisted optimization. These findings highlight the potential of coordinated RIS-UAV optimization for future 6G-enabled wireless networks, including smart-city and Internet of Things (IoT) applications.