陈复扬-自适应控制作业.docVIP

Adaptive Control and Parameter Estimation Problem 2.1 Analyze the stability of the system ， (use a Lyapunov function candidate ). Answer: V(x) 0, 0 According to Lyapunov stability theory, it is asymptotically stable. Use the simulink，x1=-2.7，x2=4.3， Oscilloscope： Problem 3.2 As an application of the result in Problem 3.1, let us consider the stability of the system Condition：are continuous. is uniformly asymptotically stable. . . Put g(x,t) as input to the system, the State response: The system is not uniformly asymptotically stable. Problem 4.2 (i)Design a least squares estimator for the parameters； (ii)Design a gradient estimator for the parameters; (iii)Show that for,the estimated parameters converge to the true ones as t goes to in?nity for any initial estimates (hint: use the de?nition of persistently exciting signals to show that in this case the signalis persistently exciting); (iv)Simulate both the gradient and least squares estimators with=1, =2 for (a) and (b) , and plot the parameter errors for ; (v) Comment on the simulation results. Answer: Estimation error: (i)least squares estimator Where are all design parameters. (ii)gradient estimator The steepest descent direction of is ,,which suggests the adaptive update law for : Where is a gain matrix withbeing a design parameter. (iii) The system input has at least n+m+1(=2) frequencies , so the signal is persistently exciting. The estimated parameters converge to the true ones. (iv) =1, =2 LSE (a) u(t)=sin(t) result = 2.0000 1.0000 fig1 u(t)=sin(t) Simulate the least squares estimators (b) u(t)=1 result =2.0000 1.0542 fig2 u(t)=1 Simulate the least squares estimators RGC (a) u(t)=sin(t) result = 2.0000 0.9621 fig1 u(t)=sin(t) Simulate the gradient estimators (b) u(t)=1 result = 2.0000 0.9878 fig2 u(t)=1 Simulate