A Genetic Algorithm-Based Neuro-Fuzzy Controller for Unmanned Aerial Vehicle Control

Boumediene Selma, Samira Chouraqui, Belkacem Selma

OAI: oai:igi-global.com:292505 • DOI: 10.4018/IJAMC.292505

In this paper a self-tuning Adaptive Neuro-fuzzy Inference System (ANFIS) Controller by Genetic Algorithm (GA) applied to trajectory tracking task of Unmanned Aerial Vehicle (UAV) is studied. The quadrotor was chosen due to its simple mechanical structure; nevertheless, these types of aircraft are highly nonlinear. A model of a non-linear closed-loop dynamic model of three degrees of freedom (3-DOF) quadrotor is developed and implemented. Intelligent control such as fuzzy logic is a suitable choice for controlling nonlinear systems. The ANFIS Controller is used to reproduce the desired trajectory of the quadrotor in 2-D Vertical plane and the GA algorithm aims is to facilitate convergence to the ANFIS’s optimal parameters in order to reduce learning errors and improve the quality of the controller. The performance of the ANFIS-GA controller is compared with a ANFIS and a conventional PID controller. Simulation results confirm the advantages of the proposed controller and approve better performance.

Algorithms Computer Science and Information Technology Artificial Intelligence