Energy-Efficient Drone Battery Management Using AI-Based Predictive Control and Smart Charging Strategies

Abstract

The growing use of unmanned aerial vehicles (UAVs) in surveillance, agriculture, transportation, and environmental monitoring has increased the demand for reliable and energy-efficient battery systems. Traditional drone batteries often suffer from limited endurance, unpredictable power consumption, and inefficient charging practices, which reduce flight duration and overall mission performance. This study proposes an AI-based predictive control system combined with smart charging strategies to enhance drone battery efficiency, safety, and lifespan. The proposed framework uses machine learning algorithms to predict key battery parameters such as state-of-charge (SoC), state-of-health (SoH), discharge patterns, and thermal variations. By analyzing real-time sensor data and historical usage profiles, the system optimizes power allocation during flight and minimizes energy losses. The predictive model adjusts the drone’s power consumption based on mission requirements, environmental conditions, and remaining energy levels, ensuring higher operational reliability. In addition, a smart charging strategy is designed using adaptive current and voltage control. This method prevents overcharging, reduces thermal stress, and maintains battery cell balance. The charging process dynamically adjusts based on predicted battery health, thereby extending overall battery lifecycle and ensuring faster, safer charging cycles Simulation results indicate improved flight time, better energy utilization, and enhanced thermal stability compared to conventional BMS approaches. The combined predictive and adaptive approach strengthens mission planning by offering accurate battery performance forecasts, accuracy of 93.45%, sensitivity of 96.52%, Recall of 97.23% has attained.