PID Tuning Of Quadrotor Attitude Control Based On Mended Firefly Algorithm

Quadrotor unmanned aerial vehicle(UAV)is widely used in various fields because of its superior performance.Its flight control system is a typical underactuated,statically unstable system with high coupling,nonlinear,time-varying and other characteristics.It is a relatively complex control object,and the control system design is difficult.How to design an accurate aircraft pose controller has become an urgent problem to be solved.On the basis of understanding the composition of the control system of the quadrotor aircraft,a dynamic model of the underactuated system with six degrees of freedom and four control inputs is built to study the design method of the controller.PID control is the most basic and common control method,the controller has a simple structure,high system stability,but the traditional PID controller parameter setting method such as test method and the method of Z-N,it is difficult to make the PID algorithm to achieve the best control effect,and the inherent parameters of four rotor dynamics model change cannot adjust PID parameters in real time,In other words,when the dynamics model of quadrotor changes,the robustness of PID-based pose control system cannot meet the actual requirements.In this paper,firefly algorithm is used to self-tune PID parameters.Aiming at the problems of slow convergence speed and low accuracy of basic Firefly algorithm(FA),linear decreasing step factor,inertia weight and adjusting factor are introduced.Meanwhile,in order to eliminate the influence of random distribution of initial firefly population on firefly algorithm performance,An improved firefly control algorithm(MFA)was developed by using a method strategy combining homogenization and randomization to generate the initial firefly population.And on the basis of considering several main time domain performance indexes comprehensively,the comprehensive performance indexes are proposed as fitness function of MFA algorithm.By changing the weight coefficient,the MFA-PID controller can be made to be more specific to one or some specified performance parameters effectively,that is,the four-rotor PID controller can be adjusted for different occasions.In addition,the integral term of PID controller will accumulate deviation,resulting in the integral saturation of the actuator,reducing the PID control effect.In order to solve this problem,an anti-integral saturation PID algorithm is introduced in the MFA-PID quadrotor control system.Simulation analysis of the established model,the simulation platform used for MATLAB/Simulink.Compared with Z-N method and FA,the experimental results show that the MFA has obvious improvement on the control response performance of the quadrotor aircraft: faster response,less overshoot,shorter time to reach stability;When the model parameters of the aircraft change,MFA-PID algorithm has shorter recovery time and better real-time tracking performance.The results show that the control performance of MFA-PID is better than that of other three population intelligent tuning methods.Figure [48] Table [6] Reference [98]...