Semantic-Twin-Enabled Bifurcated Control: A Remote Control Approach for Multi-UAV Tasks

In this paper, we propose a semantic-twin-enabled bifurcated control architecture for multi-unmanned aerial vehicle (UAV) tasks in remote areas. To mitigate the effects of interference channels and construct a task-oriented digital twin (DT), we propose the concept of semantic twin (ST), which transmits semantics extracted from perceptual data and optimizes channel utilization. To achieve efficient remote control, we develop a bifurcated control architecture based on ST, in which satellites and the ground control station (GCS) function as edge controllers and the remote controller, respectively. On this basis, we design a ST-enabled model-based offline reinforcement learning algorithm for lifelong learning. By building a virtual target and refining the weighted sampling, model ensemble and regularization methods, we ensure the reliability of the virtual environment and the efficacy of offline model training. Finally, we validate this framework by designing a multi-UAV tracking task, and verify the significant advantages of the proposed control architecture in scenario reconstruction and decision performance, proving the effectiveness of ST-enabled bifurcated control architecture in remote UAV tasks.