A real-time strain monitoring framework for UAV wings based on a dual-driven approach integrating physical models and data

Abstract

With the increasing prevalence of drone applications, monitoring their health status has become crucial. This paper compares the prediction accuracy and spatial range of predictions between data-driven and dual-driven methods, which combine models and data. A real-time strain monitoring framework for UAV wings based on a dual-driven approach integrating physical models and data is proposed. This framework overcomes the limitations of single-method approaches and keeps prediction errors within 10%. Additionally, by utilizing real-time data from sensors and visualizing the predicted wing model, real-time monitoring of UAV wing strain is achieved. This provides a robust safeguard for the safe flight and mission execution of UAVs.