Marhoon, Hamzah M. and Omar, Rasha Khalid and Al-Rammahi, Hussein and Al-Tahir, Sarah O. and Basil, Noorulden and Tarik, Benmessaoud Mohammed and Agajie, Takele Ferede (2025) Metaheuristic-Driven Optimisation of Support Vector Regression Models for Precision Control in Unmanned Aerial Vehicle Systems. Buletin Ilmiah Sarjana Teknik Elektro, 7 (3). pp. 608-624.

Text 14251-Article Text-64643-1-10-20251007.pdf Download (1MB)

Abstract

Unmanned Aerial Vehicle (UAV) systems are deployed in dynamic and uncertain environments where many traditional control structures, including Proportional–Integral–Derivative (PID) and Linear Quadratic Regulator (LQR) controllers, are unable to provide stability and adaptation. In order to overcome these shortcomings, this work presents a hybrid Support Vector Regression (SVR) model optimised with the Eagle Strategy-Particle Swarm Optimisation (ES-PSO). The proposed framework is tested with high-fidelity simulated flight data on a quadcopter platform, in which throttle, pitch, roll and yaw are provided as control variables and altitude, velocity and orientation are provided as outputs. The ES-PSO algorithm is an algorithm that optimises the global and local hyperparameters of the SVR and makes it more effective at capturing nonlinear dynamics of the input-output process under both nominal and perturbed flight conditions. To compare with benchmarking, standalone SVR, Neural Networks, Decision Trees, Naive Bayes and K-Nearest Neighbour models were executed using the same simulation parameters with no metaheuristic optimisation, and it was made fair. Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Percentage Error (MPE) quantitative assessments illustrate that the ES-PSO-SVR model has the lowest error in prediction and the highest tracking accuracy compared to all baseline techniques. These results demonstrate how metaheuristic-based learning systems can be used to drive forward the creation of adaptive and intelligent UAV control systems that can perform effectively in challenging operational conditions.

Item Type:	Artikel Umum
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi:	Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Electrical Engineering (S1-Teknik Elektro)
Depositing User:	M.Eng. Alfian Ma'arif
Date Deposited:	03 Nov 2025 04:08
Last Modified:	03 Nov 2025 04:08
URI:	http://eprints.uad.ac.id/id/eprint/88423
Dosen Pembimbing:	UNSPECIFIED | [error in script]

Actions (login required)

View Item