Mogorosi, Tony Oliver (Botswana International University of Science and Technology), Jamisola, Rodrigo S. Jr. (Botswana International University of Science and Technology), Subaschandar, N. (Botswana International University of Science and Technology), Mohutsiwa, Lucky Odirile (Botswana International University of Science and Technology)

Thrust­-To-­Weight Ratio Optimization for Multi-­Rotor Drones Using Neural Network with Six Input Parameters

Scheduled for presentation during the Regular Session "Neural Networks" (FrB3), Friday, June 18, 2021, 11:20−11:40, Edessa

2021 International Conference on Unmanned Aircraft Systems (ICUAS), June 15-18, 2021, Athens, Greece

This information is tentative and subject to change. Compiled on April 19, 2024

Abstract

This study analyzes the thrust­-to­-weight ratio of a multi­rotor drones with respect to six different parameters using neural network. The parameters are the model weight, number of propellers, frame size, propeller diameter, propeller pitch and number of blades. An online calculation tool called eCalc is used to collect data to build a neural network model.The model has an accuracy of 97% when compared to an eCalc computed data. From this model, we optimize the thrust­-to-­weight ratio using gradient descent method initialized from the collected eCalc data. We ran another optimization computationby fixing two parameters to satisfy available components in the market. Optimization results are showed and analyzed.