Maximizing energy efficiency in drones through accurate state of charge estimation using extended Kalman filter

Abstract

This paper delves into the critical aspect of managing energy consumption in drone operations to achieve the utmost range and ensure accurate state of charge (SoC) estimation. Effective energy management is pivotal in determining the operational range of drones, allowing for longer distances and heavier payloads. The integration of precise energy estimation algorithms into operational planning extends the range of drones, facilitating swift, environmentally-conscious missions for sustainable and efficient logistics solutions. The paper introduces a mathematical model to understand energy consumption and battery behavior in drones, utilizing the hybrid pulse power characterization test and recursive least square with forgetting factor for parameter identification. To overcome the limitations of linear filters, the paper employs the accurate extended Kalman filter (EKF) in the nonlinear filter section. The EKF significantly enhances the battery management system by furnishing precise SoC data. The study evaluates two SoC estimation techniques: SoC-AH (ampere-hours) and SoC_EKF, using root mean square error for comparison. The SoC_EKF technique demonstrates higher accuracy, boasting a lower errors value of 0.78%, thus making it superior for precise drone battery SoC estimation. These findings contribute to the improved performance, reliability, and overall safety of drones.