Advanced Battery Health Management in Electric UAVs Through Digital Twins

The advancement of digital twin technology indicates new opportunities for enhancing predictive models across various sectors. For electric Unmanned Aerial Vehicles (UAVs), it is paramount to predict battery end of discharge (EOD) with precision. This study introduces a novel methodology employing digital twins to refine EOD prediction by categorizing UAV operation into multiple phases, based on throttle usage. This categorization effectively correlates voltage data with distinct operational stages, yielding more precise and situational-aware EOD forecasts. Our solution offers better accuracy than conventional approaches while simplifying the prediction process and accommodating the changing usage patterns of UAVs. This research marks a significant step towards more sophisticated battery management systems, avoiding risk operations with low battery levels and ensuring flight safety. It illustrates how digital twins could improve battery health monitoring and management for electric UAVs.