New Designing Approaches for Quadcopter PID Controllers Using Reference Model and Genetic Algorithm Techniques

Abstract

This paper aims to describe the design of two different controllers for a quadcopter. Firstly, a dynamic model of the aerial vehicle obtained via Newton Euler formalism is presented. Then a control strategy based on the reference model method is used to synthesize PID controller. This controller is tuned by using a simplified model of the quadcopter, without going through the step of the linearization. Secondly, the genetic algorithm optimization technique is employed to compute the parameters of the second PID controller. The performance is investigated for various fitness expressions and tested taking into account the presence of disturbances. For a large control domain, the both controllers are tuned independently from the flight domain. Finally, simulation results with the nonlinear quadcopter system are depicted. The obtained results show, that the proposed controllers provide efficiency in terms of stability, flight quality and disturbance rejection.