自适应控制作业一.docxVIP

Adaptive ControlAssignment 1System IdentificationDue Date16th December 2015NotesMatlab is the suggested program for this assignment, although you may use other programming languages. Your program codes are also required when you submit your assignment report.Taskse(t)+An ARX system model is depicted in Figure 1, where noise e(t) is a white noise sequence with variance 1.y(t)u(t)Figure 1 An ARX ModelObtain the system model equation and write it in linear regression form. Lecture 2, Page19, 4.Auto Regressive with eXogenous input (ARX ) Model;Page40, EXAMPLESimulate the system by generating 1000 data points. The inputs u(t) is assumed to be a uniformly distributed random signal in [-2,2]. Plot u(t) and y(t).Based on the input and output data, and the system model, obtain the least squares estimator for this system, that is, the estimation of system parameters. a)b) 上图是运行程序task1的结果c) 运行程序task1经过1000步迭代后，结果是左边是估计值，右边是真实值Task1代码，也可以直接运行附件中的task1.mN = 1000;n = 0:50; u = -2 + (2+2)*rand(N,1);e = randn(N,1); A = [1 -0.8];B = [0 1];m0 = idpoly(A,B);%??êμ?μu = iddata([],u);e = iddata([],e);y = sim(m0,[u e]);z = [y,u];m = arx(z,[1 1 1]); %×?D??t3?·¨μ?1à???μ A_est = m.A;B_est = m.B;m_est = idpoly(A_est,B_est);y_est = sim(m_est,u); figure plot(y,r)%hold onfigureplot(y_est,b) h = impz(B,A,n);h_est = impz(B_est,A_est,n); figureplot(h,r)% hold onfigureplot(h_est,b)For an ARX model, the system parameters are given as:A= [1 -1.5363 0.8607]; B=[0 0.0416 0.0395].Generate any input (eg step, sampled sin wave, random) and get the model response to this. Plot u(t) and y(t). Ignore the system noise in this case.Write a recursive least squares program to identify this model by the obtained input and output data. Test your recursive least squares program to see if it is predicting the correct system parameters.Test the response and the recursive least squares program if a white noise is added to the output.You should initialize your model with and, where LN is a large number. You