State of Charge Estimation for 18650 Lithium-Ion Batteries: A Comparative Analysis Using Technical Specifications from Three Leading Manufacturers

This paper addresses the crucial challenge of optimizing battery management for electric Unmanned Aerial Vehicles (UAVs), with a specific focus on 18650 lithium-ion (Li-Ion) batteries. Drawing upon extensive knowledge and insights gained from experimental tests and real-world missions, we aim to establish a foundational understanding of their application in UAVs and address inventory issues related to State of Charge (SoC) battery estimation. To enhance our analysis, we delve into the technical specifications of leading Li-Ion cell manufacturers—LG, Samsung, and Panasonic—to extract discharge behavior data, forming our training dataset. Employing the Four Parameter Logistic (4PL) regression model, we then introduce three distinct models, each tailored to one of the three manufacturers’ cells, demonstrating their capability to accurately estimate the SoC. To validate our approach, we apply these models to a real-world flight log, achieving satisfactory results.