Location
Online
Event Website
https://hicss.hawaii.edu/
Start Date
3-1-2023 12:00 AM
End Date
7-1-2023 12:00 AM
Description
Unmanned aerial vehicles are becoming integrated into a wide range of modern IoT and CPS environments for various industrial, military, and entertainment applications. With growing estimations for this market in the future, the problem of energy consumption and its prediction is becoming increasingly important for optimal battery-saving, as well as the safety of the application and thus protection of surrounding persons near the drone flight. This paper presents a machine learning-based approach for the prediction of the power consumption of unmanned aerial vehicles at certain times of the flight. Instead of predicting the power consumption in prescribed environments with complex, time-consuming measurement techniques, our approach is fast, easy to implement, and predicts real-world power consumption in five classes, with a balanced accuracy of 66.7 percent.
Recommended Citation
Gatscher, Julian; Breitenbach, Johannes; and Buettner, Ricardo, "Machine Learning-Based Power Consumption Prediction for Unmanned Aerial Vehicles in Dynamic Environments" (2023). Hawaii International Conference on System Sciences 2023 (HICSS-56). 5.
https://aisel.aisnet.org/hicss-56/st/mobile_app_development/5
Machine Learning-Based Power Consumption Prediction for Unmanned Aerial Vehicles in Dynamic Environments
Online
Unmanned aerial vehicles are becoming integrated into a wide range of modern IoT and CPS environments for various industrial, military, and entertainment applications. With growing estimations for this market in the future, the problem of energy consumption and its prediction is becoming increasingly important for optimal battery-saving, as well as the safety of the application and thus protection of surrounding persons near the drone flight. This paper presents a machine learning-based approach for the prediction of the power consumption of unmanned aerial vehicles at certain times of the flight. Instead of predicting the power consumption in prescribed environments with complex, time-consuming measurement techniques, our approach is fast, easy to implement, and predicts real-world power consumption in five classes, with a balanced accuracy of 66.7 percent.
https://aisel.aisnet.org/hicss-56/st/mobile_app_development/5