Performance enhancement of drone LiB state of charge using extended Kalman filter algorithm

Abstract

This study introduces a more accurate approach to managing drone batteries by improving how the state of charge (SoC) is estimated, focusing on energy efficiency and environmental impact. The key innovation lies in developing a mathematical model to assess battery behavior, combined with Hybrid Pulse Power Characterization testing and Recursive Least Squares with Forgetting Factor for parameter identification. To enhance the battery management system, the study integrates the Extended Kalman Filter (EKF), which overcomes the limitations of traditional linear filters and provides more precise SoC estimation. This approach reduces energy waste and extends battery life, directly supporting sustainable engineering practices. A developed MATLAB-based framework ensures real-time monitoring and optimized battery performance, minimizing the risk of power depletion during flight. The results demonstrate that the proposed SoC_EKF method significantly outperforms the conventional SoC_AH approach, achieving a lower estimation error (1.93 × 10 −4 vs. 7.21 × 10 −4 ), leading to improved energy efficiency, reduced carbon footprint, and more reliable, eco-friendly drone operations for clean technology applications.