Cooperative Digital Twins for UAV-Based Scenarios

The advancement of 6G wireless communication technology and the rise of the Internet of things (IoT) has made digital twin (DT) a promising tool for unmanned aerial vehicles (UAVs)-based 6G applications such as intelligent logistics, management of smart cities, and autonomous driving. DT allows UAVs to imitate the status of physical entities in a virtual space for service provisions in the physical space. However, traditional DT solutions are challenging to accurately imitate and derive highly dynamic physical entities due to the limited computing resources of UAVs. To address this issue, we propose a cooperative DT framework to achieve a highly accurate and low-latency DT performance with a double-scale spatial DT manner. We first propose a UAV-based cooperative sensing algorithm to implement comprehensive data collection for building accurate small-scale spatial DT models. Then we propose an adaptive model parameter adjustment algorithm to improve the DT accuracy. Considering the high mobility of physical entities, we propose a model transfer algorithm to achieve a low-latency service provision. Finally, we demonstrate the effectiveness of our proposed framework using a case study of UAV-based multi-target tracking. The experiment results show that our solution achieves an accurate DT with a successful tracking ratio of up to 95 percent with a low system latency of under 1 second compared to traditional DT manners on average, respectively.