A Comparative Study of Machine Learning and Deep Learning Models for State of Charge and Remaining Useful Life Estimation on a Rotary-Wing UAV Battery
DRONES, cilt.10, sa.7, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 10 Sayı: 7
- Basım Tarihi: 2026
- Doi Numarası: 10.3390/drones10070549
- Dergi Adı: DRONES
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Compendex, INSPEC, Directory of Open Access Journals, Technology Collection (ProQuest)
- Anahtar Kelimeler: battery management, gradient boosted trees, LSTM, machine learning, remaining useful life, rotary-wing UAV, RWKV, state of charge, unmanned aerial vehicle
- Erciyes Üniversitesi Adresli: Evet
Özet
Battery state estimation is the main safety constraint for electric rotary-wing unmanned aerial vehicles (UAVs): mission decisions depend on both the instantaneous State of Charge (SOC) and the Remaining Useful Life (RUL). The present study compares seven machine learning and deep learning models (LR, SVM, k-NN, GBT, EL, LSTM, and a simplified RWKV) on real flight data from a rotary-wing helicopter testbed with a Pixhawk autopilot and an NVIDIA Jetson Nano mission computer. The dataset has 1310 samples (similar to 262 s) of nine on-board sensor signals. Mission-based RUL is defined as the projected time until SOC reaches a safe-landing threshold. All models use an 80/20 random split, five regression metrics (RMSE, MAE, , MSE, PRMSE), and five random seeds. GBT wins on SOC with , MAE , and s per-sample inference on a workstation CPU; this latency leaves headroom for on-board mission planning. Battery temperature and voltage together carry over of the predictive signal. GBT wins again on RUL (, MAE s). The same ordering (tree ensemble > recurrent > linear) holds for both tasks; the remaining RUL gap reflects the single-flight dataset. The SOC labels originate from the on-board autopilot's Coulomb-counting-based fuel-gauge estimator, so the SOC numbers should be read as a reproduction of that on-board trace at sub-microsecond inference latency rather than as independent accuracy; the calibration-free Coulomb-counting baseline reaches a marginally higher (, MAE ) on the same task.