This paper presents a novel approach for joint Terahertz (THz) communication and three-dimensional (3D) map reconstruction using an Unmanned Aerial Vehicle (UAV). Due to the need for precise mapping to provide Line-of-Sight (LoS) THz communication services, our approach leverages the UAV's trajectory to initiate the reconstruction and updating of the 3D environment in real-time. The UAV starts without any prior knowledge of the target area, establishing THz communications with users scattered among 3D obstacles. A comprehensive system model is developed, incorporating realistic antenna patterns, UAV oscillations, and received signal power modeling for accurate analysis. We establish an adaptive methodology for 3D environment modeling, dynamically constructing and refining a gridbased obstacle map. A dynamic algorithm for real-time 3D map reconstruction is implemented, utilizing received signal strengths to update the environmental model. Detailed simulations validate the effectiveness of our proposed methods, demonstrating significant improvements in computational efficiency and realtime decision-making in UAV-assisted communication networks. Our results highlight the potential of this joint approach in enhancing operational efficiency and communication reliability under varying environmental conditions.
