EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Energy Consumption Prediction Model for UAV Logistics Delivery: A Study Based on Multi-Feature Fusion and Ensemble Learning

Sulin Mao

International Journal of Applied Science · 02.06.2026 · 9 (2) · p41

DOI: 10.30560/ijas.v9n2p41 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Applied Science, online erschienen am 02.06.2026. Verfasst von Sulin Mao. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent years, the low-altitude economy has developed rapidly, and drones have been increasingly used in the field of logistics delivery. The energy consumption level of drones directly affects whether delivery tasks can be successfully completed and also impacts operating costs. Therefore, studying energy consumption prediction methods has practical significance. This paper constructs a drone delivery dataset based on real flight data. We select operating parameters such as flight distance, speed, altitude, and wind speed, and design derived features like cumulative distance and flight time to build a multi-feature fusion data system. Based on this, we use four methods—Linear Regression, Random Forest, Gradient Boosting, and XGBoost—to predict drone energy consumption. We evaluate each model's performance using Mean Absolute Error, Root Mean Square Error, Coefficient of Determination, and Median Absolute Percentage Error. Experimental results show that the Random Forest model achieves the best prediction performance, with a Coefficient of Determination of 0.9985 and a Median Absolute Percentage Error of only 0.22%, significantly outperforming the Linear Regression model. Feature importance analysis shows that cumulative flight distance has the greatest impact on energy consumption, with a contribution rate of 85.13%. The results provide a quantitative basis for mission feasibility assessment, battery configuration optimization, and low-energy path planning in logistics drone delivery.

Zitieren

APA

Mao, S. (2026). Energy Consumption Prediction Model for UAV Logistics Delivery: A Study Based on Multi-Feature Fusion and Ensemble Learning. International Journal of Applied Science, 9(2), p41. https://doi.org/10.30560/ijas.v9n2p41

Vancouver

Mao S. Energy Consumption Prediction Model for UAV Logistics Delivery: A Study Based on Multi-Feature Fusion and Ensemble Learning. International Journal of Applied Science. 2026;9(2):p41. doi:10.30560/ijas.v9n2p41

BibTeX
@article{Mao2026,
  author = {Mao, Sulin},
  title = {Energy Consumption Prediction Model for UAV Logistics Delivery: A Study Based on Multi-Feature Fusion and Ensemble Learning},
  journal = {International Journal of Applied Science},
  year = {2026},
  volume = {9},
  number = {2},
  pages = {p41},
  doi = {10.30560/ijas.v9n2p41}
}
Kostenlos starten →