Nonlinear Control Lecture # 20 Special nonlinear Forms Nonlinear Control Lecture # 20 Special nonlinear Forms
Normal Form Relative Degree x = f ( x ) + g ( x ) u, ˙ y = h ( x ) where f , g , and h are sufficiently smooth in a domain D f : D → R n and g : D → R n are called vector fields on D y = ∂h def ˙ ∂x [ f ( x ) + g ( x ) u ] = L f h ( x ) + L g h ( x ) u L f h ( x ) = ∂h ∂xf ( x ) is the Lie Derivative of h with respect to f or along f Nonlinear Control Lecture # 20 Special nonlinear Forms
L g L f h ( x ) = ∂ ( L f h ) g ( x ) ∂x f h ( x ) = L f L f h ( x ) = ∂ ( L f h ) L 2 f ( x ) ∂x h ( x ) = ∂ ( L k − 1 h ) f h ( x ) = L f L k − 1 f L k f ( x ) f ∂x L 0 f h ( x ) = h ( x ) y = L f h ( x ) + L g h ( x ) u ˙ L g h ( x ) = 0 ⇒ y = L f h ( x ) ˙ y (2) = ∂ ( L f h ) [ f ( x ) + g ( x ) u ] = L 2 f h ( x ) + L g L f h ( x ) u ∂x Nonlinear Control Lecture # 20 Special nonlinear Forms
y (2) = L 2 L g L f h ( x ) = 0 ⇒ f h ( x ) y (3) = L 3 f h ( x ) + L g L 2 f h ( x ) u L g L ρ − 1 L g L i − 1 h ( x ) = 0 , i = 1 , 2 , . . . , ρ − 1; h ( x ) � = 0 f f y ( ρ ) = L ρ f h ( x ) + L g L ρ − 1 h ( x ) u f Definition 8.1 The system x = f ( x ) + g ( x ) u, ˙ y = h ( x ) has relative degree ρ , 1 ≤ ρ ≤ n , in R ⊂ D if ∀ x ∈ R L g L i − 1 L g L ρ − 1 h ( x ) = 0 , i = 1 , 2 , . . . , ρ − 1; h ( x ) � = 0 f f Nonlinear Control Lecture # 20 Special nonlinear Forms
Example 8.1 Controlled van der Pol equation x 2 = ε [ − x 1 + x 2 − 1 3 x 3 x 1 = x 2 /ε, ˙ ˙ 2 + u ] , y = x 1 x 2 /ε = − x 1 + x 2 − 1 3 x 3 y = ˙ ˙ x 1 = x 2 /ε, y = ˙ ¨ 2 + u Relative degree two over R 2 x 2 = ε [ − x 1 + x 2 − 1 3 x 3 x 1 = x 2 /ε, ˙ ˙ 2 + u ] , y = x 2 y = ε [ − x 1 + x 2 − 1 3 x 3 Relative degree one over R 2 ˙ 2 + u ] , x 2 = ε [ − x 1 + x 2 − 1 3 x 3 y = 1 2 ( ε 2 x 2 1 + x 2 x 1 = x 2 /ε, ˙ ˙ 2 + u ] , 2 ) y = ε 2 x 1 ˙ x 2 = εx 2 2 − ( ε/ 3) x 4 ˙ x 1 + x 2 ˙ 2 + εx 2 u Relative degree one in { x 2 � = 0 } Nonlinear Control Lecture # 20 Special nonlinear Forms
Example 8.2 (Field-controlled DC motor) x 1 ˙ = d 1 ( − x 1 − x 2 x 3 + V a ) x 2 ˙ = d 2 [ − f e ( x 2 ) + u ] x 3 ˙ = d 3 ( x 1 x 2 − bx 3 ) y = x 3 y = ˙ ˙ x 3 = d 3 ( x 1 x 2 − bx 3 ) y = d 3 ( x 1 ˙ ¨ x 2 + ˙ x 1 x 2 − b ˙ x 3 ) = ( · · · ) + d 2 d 3 x 1 u Relative degree two in { x 1 � = 0 } Nonlinear Control Lecture # 20 Special nonlinear Forms
Example 8.3 H ( s ) = b m s m + b m − 1 s m − 1 + · · · + b 0 s n + a n − 1 s n − 1 + · · · + a 0 x = Ax + Bu, ˙ y = Cx 0 1 0 0 . . . . . . 0 0 0 1 0 . . . . . . 0 . . ... . . . . . . . ... A = , B = ... . . . . . ... . . . . 0 0 0 0 1 1 − a 0 − a 1 . . . . . . − a m . . . . . . − a n − 1 � � C = b 0 b 1 . . . . . . b m 0 . . . 0 Nonlinear Control Lecture # 20 Special nonlinear Forms
y = CAx + CBu, ˙ If m = n − 1 , CB = b n − 1 � = 0 ⇒ ρ = 1 CA i − 1 B = 0 , i = 1 , . . . , n − m − 1 , CA n − m − 1 B = b m � = 0 y ( n − m ) = CA n − m x + CA n − m − 1 Bu ⇒ ρ = n − m 1 H ( s ) = N ( s ) N ( s ) Q ( s ) D ( s ) = Q ( s ) N ( s ) + R ( s ) = R ( s ) 1 1 + Q ( s ) N ( s ) u + e y ✲ ✲ 1 ✲ ❦ Q ( s ) − ✻ w R ( s ) ✛ N ( s ) Nonlinear Control Lecture # 20 Special nonlinear Forms
� y, . . . , y ( ρ − 1) � State model of 1 /Q ( s ) : ξ = col y, ˙ ˙ ξ = ( A c + B c λ T ) ξ + B c b m e, y = C c ξ 0 1 0 0 . . . 0 0 0 1 0 . . . 0 . . ... . . . . � � A c = . . , B c = , C c = 1 0 . . . 0 0 . . . 0 . 0 1 1 0 . . . . . . 0 0 State model of R ( s ) /N ( s ) η = A 0 η + B 0 y, ˙ w = C 0 η Nonlinear Control Lecture # 20 Special nonlinear Forms
State model of H ( s ) η ˙ = A 0 η + B 0 C c ξ ˙ A c ξ + B c ( λ T ξ − b m C 0 η + b m u ) ξ = y = C c ξ The eigenvalues of A 0 are the zeros of H ( s ) Nonlinear Control Lecture # 20 Special nonlinear Forms
Change of variables: φ 1 ( x ) . . . φ n − ρ ( x ) φ ( x ) η def def z = T ( x ) = − − − = − − − = − − − h ( x ) ψ ( x ) ξ . . . L ρ − 1 h ( x ) f φ 1 to φ n − ρ are chosen such that T ( x ) is a diffeomorphism on a domain D x ⊂ R When ρ = n, z = T ( x ) = ψ ( x ) = ξ Nonlinear Control Lecture # 20 Special nonlinear Forms
∂φ η ˙ = ∂x [ f ( x ) + g ( x ) u ] = f 0 ( η, ξ ) + g 0 ( η, ξ ) u ˙ ξ i = ξ i +1 , 1 ≤ i ≤ ρ − 1 ˙ L ρ f h ( x ) + L g L ρ − 1 ξ ρ = h ( x ) u f y = ξ 1 Choose φ ( x ) such that T ( x ) is a diffeomorphism and ∂φ i ∂x g ( x ) = 0 , for 1 ≤ i ≤ n − ρ, ∀ x ∈ D x Always possible (at least locally) η = f 0 ( η, ξ ) ˙ Nonlinear Control Lecture # 20 Special nonlinear Forms
Theorem 8.1 Suppose the system x = f ( x ) + g ( x ) u, ˙ y = h ( x ) has relative degree ρ ( ≤ n ) in R . If ρ = n , then for every x 0 ∈ R , a neighborhood N of x 0 exists such that the map T ( x ) = ψ ( x ) , restricted to N , is a diffeomorphism on N . If ρ < n , then, for every x 0 ∈ R , a neighborhood N of x 0 and smooth functions φ 1 ( x ) , . . . , φ n − ρ ( x ) exist such that ∂φ i ∂x g ( x ) = 0 , for 1 ≤ i ≤ n − ρ � φ ( x ) � is satisfied for all x ∈ N and the map T ( x ) = , ψ ( x ) restricted to N , is a diffeomorphism on N Nonlinear Control Lecture # 20 Special nonlinear Forms
Normal Form: η ˙ = f 0 ( η, ξ ) ˙ ξ i = ξ i +1 , 1 ≤ i ≤ ρ − 1 ˙ L ρ f h ( x ) + L g L ρ − 1 ξ ρ = h ( x ) u f y = ξ 1 0 1 0 . . . 0 0 0 0 1 . . . 0 0 . . ... . . . . A c = . . , B c = . . . 0 . 0 1 1 0 . . . . . . 0 0 � 1 0 � C c = 0 . . . 0 Nonlinear Control Lecture # 20 Special nonlinear Forms
η ˙ = f 0 ( η, ξ ) ˙ f h ( x ) + L g L ρ − 1 L ρ � � ξ = A c ξ + B c h ( x ) u f y = C c ξ ˜ γ ( η, ξ ) = L g L ρ − 1 ψ ( η, ξ ) = L ρ � � f h ( x ) x = T − 1 ( z ) , ˜ h ( x ) � � f x = T − 1 ( z ) ξ = A c ξ + B c [ ˜ ˙ ψ ( η, ξ ) + ˜ γ ( η, ξ ) u ] If x ∗ is an open-loop equilibrium point at which y = 0 ; i.e., f ( x ∗ ) = 0 and h ( x ∗ ) = 0 , then ψ ( x ∗ ) = 0 . Take φ ( x ∗ ) = 0 so that z = 0 is an open-loop equilibrium point. Nonlinear Control Lecture # 20 Special nonlinear Forms
Zero Dynamics η ˙ = f 0 ( η, ξ ) ˙ L ρ f h ( x ) + L g L ρ − 1 � � ξ = A c ξ + B c h ( x ) u f y = C c ξ L ρ f h ( x ( t )) y ( t ) ≡ 0 ⇒ ξ ( t ) ≡ 0 ⇒ u ( t ) ≡ − L g L ρ − 1 h ( x ( t )) f ⇒ ˙ η = f 0 ( η, 0) Definition The equation ˙ η = f 0 ( η, 0) is called the zero dynamics of the system. The system is said to be minimum phase if the zero dynamics have an asymptotically stable equilibrium point in the domain of interest (at the origin if T (0) = 0 ) Nonlinear Control Lecture # 20 Special nonlinear Forms
Z ∗ = { x ∈ R | h ( x ) = L f h ( x ) = · · · = L ρ − 1 h ( x ) = 0 } f y ( t ) ≡ 0 ⇒ x ( t ) ∈ Z ∗ � L ρ f h ( x ) � def ⇒ u = u ∗ ( x ) = − � L g L ρ − 1 h ( x ) � f � x ∈ Z ∗ The restricted motion of the system is described by � L ρ � f h ( x ) def x = f ∗ ( x ) ˙ = f ( x ) − g ( x ) L g L ρ − 1 h ( x ) f x ∈ Z ∗ Nonlinear Control Lecture # 20 Special nonlinear Forms
Example 8.4 x 2 = ε [ − x 1 + x 2 − 1 3 x 3 x 1 = x 2 /ε, ˙ ˙ 2 + u ] , y = x 2 x 2 = ε [ − x 1 + x 2 − 1 3 x 3 y = ˙ ˙ 2 + u ] ⇒ ρ = 1 The system is in the normal form with η = x 1 and ξ = x 2 y ( t ) ≡ 0 ⇒ x 2 ( t ) ≡ 0 ⇒ ˙ x 1 = 0 Non-minimum phase Nonlinear Control Lecture # 20 Special nonlinear Forms
Example 8.5 x 1 = − x 1 + 2 + x 2 3 ˙ u, ˙ x 2 = x 3 , ˙ x 3 = x 1 x 3 + u, y = x 2 1 + x 2 3 y = ˙ ˙ x 2 = x 3 y = ˙ ¨ x 3 = x 1 x 3 + u ⇒ ρ = 2 Z ∗ = { x 2 = x 3 = 0 } u = u ∗ ( x ) = 0 ⇒ x 1 = − x 1 ˙ Minimum phase Nonlinear Control Lecture # 20 Special nonlinear Forms
Find φ ( x ) such that 2+ x 2 3 ∂φ 1+ x 2 � � 3 ∂φ ∂φ ∂φ φ (0) = 0 , ∂xg ( x ) = ∂x 1 , ∂x 2 , = 0 0 ∂x 3 1 and � φ ( x ) � T T ( x ) = x 2 x 3 is a diffeomorphism · 2 + x 2 ∂φ + ∂φ 3 = 0 1 + x 2 ∂x 1 ∂x 3 3 φ ( x ) = x 1 − x 3 − tan − 1 x 3 Nonlinear Control Lecture # 20 Special nonlinear Forms
Recommend
More recommend