Different Intelligent Methods for Coefficient Tuning of Quadrotor Feedback-linearization Controller

This paper investigates different intelligent methods of tuning feedback-linearization control coefficients. Feedback-linearization technique is an effective method of controlling nonlinear systems. The most critical part of designing this controller is tuning the gains, especially if the plant has complex nonlinear dynamics. In this research, to improve the performance of the overall closed-loop system, the feedback linearization method has been integrated with the conventional proportional-integral-derivative (PID) controller. Also, a quadratic performance index was used to compare the functionality of the controllers tuned by the proposed intelligent methods. These intelligent methods include Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Fuzzy Logic, and Neural Network tuning algorithms. A quadrotor aircraft is used as the plant under study in order to evaluate the performance of the controllers tunned in this research. Finally, MATLAB simulation tests demonstrate the effectiveness of the presented algorithms. According to the results, it is demonstrated that the class of online algorithms performs better, even with the specified perturbation.

Keywords:

Feedback-linearization , GA algorithm , PSO algorithm , Fuzzy-logic , Neural-network