1 Relative degree and local normal forms

Size: px
Start display at page:

Download "1 Relative degree and local normal forms"

Transcription

1 THE ZERO DYNAMICS OF A NONLINEAR SYSTEM 1 Relative degree and local normal orms The purpose o this Section is to show how single-input single-output nonlinear systems can be locally given, by means o a suitable change o coordinates in the state space, a normal orm o special interest, on which several important properties can be elucidated. The point o departure o the whole analysis is the notion o relative degree o the system, which is ormally described in the ollowing way. The single-input single-output nonlinear system ẋ = (x) + g(x)u (1) y = h(x) is said to have relative degree r at a point x i 1 (i) L g L k h(x) = 0 or all x in a neighborhood o x and all k < r 1 (ii) L g L r 1 h(x ) 0. Note that there may be points where a relative degree cannot be deined. This occurs, in act, when the irst unction o the sequence L g h(x), L g L h(x),..., L g L k h(x),... which is not identically zero (in a neighborhood o x ) has a zero exactly at the point x = x. However, the set o points where a relative degree can be deined is clearly an open and dense subset o the set U where the system (1) is deined. Remark. In order to compare the notion thus introduced with a amiliar concept, let us calculate the relative degree o a linear system ẋ = Ax + Bu y = Cx. In this case, since (x) = Ax, g(x) = B, h(x) = Cx, it easily seen that and thereore L k h(x) = CAk x L g L k h(x) = CAk B. Thus, the integer r is characterized by the conditions CA k B = 0 or all k < r 1 CA r 1 B 0. 1 Let λ be real-valued unction and an n-vector-valued vector, both deined on a subset U o R n. The unction L λ is the real-valued unction deined as L λ(x) = n i=1 This unction is sometimes called derivative o λ along. λ i(x) := λ x i x (x). 1

2 It is well-known that the integer satisying these conditions is exactly equal to the dierence between the degree o the denominator polynomial and the degree o the numerator polynomial o the transer unction o the system. T (s) = C(sI A) 1 B We illustrate now a simple interpretation o the notion o relative degree, which is not restricted to the assumption o linearity considered in the previous Remark. Assume the system at some time t is in the state x(t ) = x and suppose we wish to calculate the value o the output y(t) and o its derivatives with respect to time y (k) (t), or k = 1, 2,..., at t = t. We obtain y(t ) = h(x(t )) = h(x ) y (1) (t) = h dx x dt = h ((x(t)) + g(x(t))u(t)) x = L h(x(t)) + L g h(x(t))u(t). I the relative degree r is larger than 1, or all t such that x(t) is near x, i.e. or all t near t, we have L g h(x(t)) = 0 and thereore This yields y (2) (t) = L h dx x dt y (1) (t) = L h(x(t)). = L h ((x(t)) + g(x(t))u(t)) x = L 2 h(x(t)) + L gl h(x(t))u(t). Again, i the relative degree is larger than 2, or all t near t we have L g L h(x(t)) = 0 and Continuing in this way, we get y (2) (t) = L 2 h(x(t)). y (k) (t) = L k h(x(t)) or all k < r and all t near t y (r) (t ) = L r h(x ) + L g L r 1 h(x )u(t ). Thus, the relative degree r is exactly equal to the number o times one has to dierentiate the output y(t) at time t = t in order to have the value u(t ) o the input explicitly appearing. Note also that i L g L k h(x) = 0 or all x in a neighborhood o x and all k 0 (in which case no relative degree can be deined at any point around x ) then the output o the system is not aected by the input, or all t near t. As a matter o act, i this is the case, the previous calculations show that the Taylor series expansion o y(t) at the point t = t has the orm y(t) = L k h(x ) (t t ) k k! k=0 2

3 i.e. that y(t) is a unction depending only on the initial state and not on the input. These calculations suggest that the unctions h(x), L h(x),..., L r 1 h(x) must have a special importance. As a matter o act, it is possible to show that they can be used in order to deine, at least partially, a local coordinates transormation around x (recall that x is a point where L g L r 1 h(x ) 0). This act is based on the ollowing property. Lemma 1 The row vectors 2 are linearly independent. dh(x ), dl h(x ),..., dl r 1 h(x ) Lemma 1 shows that necessarily r n and that the r unctions h(x), L h(x),..., L r 1 h(x) qualiy as a partial set o new coordinate unctions around the point x. As we shall see in a moment, the choice o these new coordinates entails a particularly simple structure or the equations describing the system. However, beore doing this, it is convenient to summarize the results discussed so ar in a ormal statement, that also illustrates a way in which the set o new coordinates can be completed in case the relative degree r is strictly less than n. Proposition 1 Suppose the system has relative degree r at x. Then r n. I r is strictly less than n, it is always possible to ind n r more unctions ψ 1 (x),..., ψ r r (x) such that the mapping Φ(x) = ψ 1 (x)... ψ n r (x) h(x) L h(x)... L r 1 has a jacobian matrix which is nonsingular at x and thereore qualiies as a local coordinates transormation in a neighborhood o x. The value at x o these additional unctions can be ixed arbitrarily. Moreover, it is always possible to choose ψ 1 (x),..., ψ n r (x) in such a way that L g ψ i (x) = 0 or all 1 i n r and all x around x. h(x) The description o the system in the new coordinates is ound very easily. Set ψ 1 (x) h(x) z = ψ 2 (x), ξ = L h(x) ψ n r (x) h(x) L r 1 2 Let λ be a real-valued unction deined on a subset U o R n. Its dierential, denote dλ(x) is the row vector ( λ dλ(x) = x 1 λ x 2 λ ) := λ x n x. 3

4 and x = (z, ξ). Looking at the calculations already carried out at the beginning, we obtain or ξ 1,..., ξ r dξ 1 dt dξ r 1 dt = h dx x dt = L h(x(t)) = ξ 2 (t) = (Lr 2 h) dx x dt = Lr 1 h(x(t)) = ξ r (t). For ξ r we obtain dξ r dt = Lr h(x(t)) + L gl r 1 h(x(t))u(t). On the right-hand side o this equation we must now replace x(t) with its expression as a unction o x(t), which will be written as x(t) = Φ 1 (z(t), ξ(t)). Thus, setting the equation in question can be rewritten as dξ r dt q(z, ξ) = L r h(φ 1 (z, ξ)) b(z, ξ) = L g L r 1 h(φ 1 (z, ξ)) = q(z(t), ξ(t)) + b(z(t), ξ(t))u(t). Note that at the point (z, ξ ) = Φ(x ), b(z, ξ ) 0 by deinition. Thus, the coeicient a(z, ξ) is nonzero or all (z, ξ) in a neighborhood o (z, ξ ). As ar as the other new coordinates are concerned, we cannot expect any special structure or the corresponding equations, i nothing else has been speciied. However, i ψ 1 (x),..., ψ n r (x) have been chosen in such a way that L g ψ i (x) = 0, then Setting dz i dt = ψ i x ((x(t)) + g(x(t))u(t)) = L ψ i (x(t)) + L g ψ i (x(t))u(t) = L ψ i (x(t)). L ψ 1 (Φ 1 (z, ξ)) 0 (z, ξ) = L ψ n r (Φ 1 (z, ξ)) the latter can be rewritten as dz dt = 0(z(t), ξ(t)). Thus, in summary, the state-space description o the system in the new coordinates will be as ollows ż = 0 (z, ξ) ξ 1 = ξ 2 ξ 2 = ξ 3 (2) ξ r 1 = ξ r ξ r = q(z, ξ) + b(z, ξ)u. 4

5 In addition to these equations one has to speciy how the output o the system is related to the new state variables. But, being y = h(x), it is immediately seen that y = ξ 1. (3) The equations thus deined are said to be in normal orm. They are useul in understanding how certain control problems can be solved. 2 The Zero Dynamics In this section we introduce and discuss an important concept, that in many instances plays a role exactly similar to that o the zeros o the transer unction in a linear system. We have already seen that the relative degree r o a linear system can be interpreted as the dierence between the number o poles and the number o zeros in the transer unction. In particular, any linear system in which r is strictly less than n has zeros in its transer unction. On the contrary, i r = n the transer unction has no zeros; thus, a nonlinear system having relative degree r = n in some sense analogue to a linear systems without zeros. We shall see in this section that this kind o analogy can be pushed much urther. Consider a nonlinear system with r strictly less than n and look at its normal orm. Recall that, i x is such that (x ) = 0 and h(x ) = 0, then necessarily the set ξ o the last r new coordinates is 0 at x. Note also that it is always possible to choose arbitrarily the value at x o the irst n r new coordinates, thus in particular being 0 at x. Thereore, without loss o generality, one can assume that ξ = 0 and z = 0 at x. Thus, i x was an equilibrium or the system in the original coordinates, its corresponding point (z, ξ) = (0, 0) is an equilibrium or the system in the new coordinates and rom this we deduce that q(0, 0) = 0 0 (0, 0) = 0. Suppose now we want to analyze the ollowing problem, called the Problem o Zeroing the Output. Find, i any, pairs consisting o an initial state x and o an input unction u ( ), deined or all t in a neighborhood o t = 0, such that the corresponding output y(t) o the system is identically zero or all t in a neighborhood o t = 0. O course, we are interested in inding all such pairs (x, u ) and not simply in the trivial pair x = 0, u = 0 (corresponding to the situation in which the system is initially at rest and no input is applied). We perorm this analysis on the normal orm o the system. Recalling that in the normal orm y(t) = ξ 1 (t), we see that the constraint y(t) = 0 or all t implies that is ξ(t) = 0 or all t. ξ 1 (t) = ξ 2 (t) =... = ξ r (t) = 0, 5

6 Thus, we see that when the output o the system is identically zero its state is constrained to evolve in such a way that also ξ(t) is identically zero. In addition, the input u(t) must necessarily be the unique solution o the equation 0 = q(z(t), 0) + b(z(t), 0)u(t) (recall that b(z(t), 0) 0 i z(t) is close to 0). As ar as the variable z(t) is concerned, it is clear that, being ξ(t) identically zero, its behavior is governed by the dierential equation ż(t) = 0 (z(t), 0). (4) From this analysis we deduce the ollowing acts. I the output y(t) has to be zero, then necessarily the initial state o the system must be set to a value such that ξ(0) = 0, whereas z(0) = z can be chosen arbitrarily. According to the value o z, the input must be set as q(z(t), 0) u(t) = b(z(t), 0) where z(t) denotes the solution o the dierential equation ż(t) = 0 (z(t), 0) with initial condition z(0) = z. Note also that or each set o initial data ξ = 0 and z = z the input thus deined is the unique input capable to keep y(t) identically zero or all times. The dynamics o (4) correspond to the dynamics describing the internal behavior o the system when input and initial conditions have been chosen in such a way as to constrain the output to remain identically zero. These dynamics, which are rather important in many o our developments, are called the zero dynamics o the system. Remark. In order to understand why we used the terminology zero dynamics in dealing with the dynamical system (4), it is convenient to examine how these dynamics are related to the zeros o the transer unction in a linear system. Let T (s) = K b 0 + b 1 s + + b n r 1 s n r 1 + s n r a 0 + a 1 s + + a n 1 s n 1 + s n denote the transer unction o a linear system (where r characterizes, as expected, the relative degree). Suppose the numerator and denominator polynomials are relatively prime and consider a minimal realization o T (s) ẋ = Ax + Bu y = Cx with A = a 0 a 1 a 2 a n 1 C = ( b 0 b 1 b n r ). 6 B = K

7 Let us now calculate its normal orm. For the last r new coordinates we know we have to take ξ 1 = Cx = b 0 x 1 + b 1 x b n r 1 x n r + x n r+1 ξ 2 = CAx = b 0 x 2 + b 1 x b n r 1 x n r+1 + x n r+2 ξ r = CA r 1 x = b 0 x r + b 1 x r b n r 1 x n 1 + x n. For the irst n r new coordinates we have some reedom o choice (provided that the conditions stated in Proposition 1 are satisied), but the simplest one is z 1 = x 1 z 2 = x 2 z n r = x n r. This is indeed an admissible choice because the corresponding coordinates transormation x = Φ(x) has a jacobian matrix with the ollowing structure Φ x = ( ) and thereore nonsingular. In the new coordinates we obtain equations in normal orm, which, because o the linearity o the system, have the ollowing structure ż = F z + Gξ ξ 1 = ξ 2 ξ 2 = ξ 3 ξ r 1 = ξ r ξ r = Hz + Kξ + bu where H and K are row vectors and F and G matrices, o suitable dimensions. The zero dynamics o this system, according to our previous deinition, are those o ż = F z. The particular choice o the irst n r new coordinates (i.e. o the elements o z) entails a particularly simple structure or the matrices F and G. As a matter o act, is easily checked 7

8 that dz 1 dt dz n r 1 dt dz n r dt = dx 1 dt = x 2(t) = z 2 (t) = dx n r 1 dt = x n r (t) = z n r (t) = dx n r = x n r+1 (t) = b 0 x 1 (t) b n r 1 x n r (t) + ξ 1 (t) dt = b 0 z 1 (t) b n r 1 z n r (t) + ξ 1 (t) rom which we deduce that F = , G = b 0 b 1 b 2 b n r From the particular orm o this matrix, it is clear that the eigenvalues o F coincide with the zeros o the numerator polynomial o T (s), i.e. with the zeros o the transer unction. Thus it is concluded that in a linear system the zero dynamics are linear dynamics with eigenvalues coinciding with the zeros o the transer unction o the system. 3 Global Normal Forms We address, in this section, the problem o deriving the global version o the coordinates transormation and normal orm introduced in section 1. Consider a single-input singleoutput system described by equations o the orm ẋ = (x) + g(x)u y = h(x) in which (x) and g(x) are smooth vector ields, and h(x) is a smooth unction, deined on R n. Assume, as usual, that (0) = 0 and h(0) = 0. This system is said to have uniorm relative degree r i it has relative degree r at each x R n. I system (5) has uniorm relative degree r, the r dierentials dh(x), dl h(x),..., dl r 1 h(x) are linearly independent at each x R n and thereore the set Z = {x R n : h(x) = L h(x) =... = L r 1 h(x) = 0} (which is nonempty in view o the hypothesis that (0) = 0 and h(0) = 0) is a smooth embedded submaniold o R n, o dimension n r. In particular, each connected component o Z is a maximal integral maniold o the (nonsingular and involutive) distribution = (span{dh, dl h,..., dl r 1 h}). The submaniold Z is the point o departure or the construction a globally deined version o the coordinates transormation considered in section 1. 8 (5)

9 Proposition 2 Suppose (5) has uniorm relative degree r. Set α(x) = Lr h(x) L g L r 1 h(x) and consider the (globally deined) vector ields β(x) = 1 L g L r 1 h(x) Suppose the vector ields (x) = (x) + g(x)α(x), g(x) = g(x)β(x). τ i = ( 1) i 1 ad i 1 g(x), 1 i r (6) are complete. Then Z is connected. Moreover, the smooth mapping Φ : Z R r R n (z, (ξ 1,..., ξ r )) Φ τ 1 ξ r Φ τ 2 ξ r 1 Φ τ r ξ 1 (z), in which as usual Φ τ t (x) denotes the low o the vector ield τ, has a globally deined smooth inverse (z, (ξ 1,..., ξ r )) = Φ 1 (x) (8) in which z = Φ τr h(x) Φτ 2 ξ i = L i 1 h(x) 1 i r. L r 2 h(x) Φτ 1 (x) L r 1 h(x) The globally deined dieomorphism (8) changes system (5) into a system described by equations o the orm (7) ż = 0 (z, ξ 1,..., ξ r ) ξ 1 = ξ 2 ξ r 1 = ξ r ξ r = q(z, ξ 1,..., ξ r ) + b(z, ξ 1,..., ξ r )u (9) where y = ξ 1 q(z, ξ 1,..., ξ r ) = L r h Φ(z, (ξ 1,..., ξ r )) b(z, ξ 1,..., ξ r ) = L g L r 1 h Φ(z, (ξ 1,..., ξ r )). I, and only i, the vector ields (6) are such that 3 [τ i, τ j ] = 0 or all 1 i, j r, 3 Let and g be vector ields on R n. Their Lie bracket, denoted [, g], is the vector ield o R n deined as [, g](x) = g (x) x x g(x). 9

10 then the globally deined dieomorphism (8) changes system (5) into a system described by equations o the orm ż = 0 (z, ξ 1 ) ξ 1 = ξ 2 ξ r 1 = ξ r ξ r = q(z, ξ 1,..., ξ r ) + b(z, ξ 1,..., ξ r )u y = ξ 1. (10) Note also that, i r < n, the submaniold Z is the largest (with respect to inclusion) smooth submaniold o h 1 (0) with the property that, at each x Z, there is u (x) such that (x) = (x) + g(x)u (x) is tangent to Z. Actually, or each x Z there is only one u (x) rendering this condition satisied, namely, u (x) = Lr h(x) L g L r 1 h(x). The submaniold Z is an invariant maniold o the (autonomous) system ẋ = (x) (11) and the restriction o this system to Z, which can be identiied with (n r)-dimensional system ż = 0 (z, 0,..., 0) o Z, is the system which characterizes the zero dynamics o (5). 4 Robust stabilization via partial-state eedback An typical setting in which normal orms are useul is a systematic method or stabilization in the large o certain classes o nonlinear system, even in the presence o parameter uncertainties. We discuss irst the case o systems having relative degree 1, i.e. we consider systems modelled by equations o the orm ż = 0 (z, ξ) ξ = q(z, ξ) + b(z, ξ)u (12) y = ξ 1 in which z R n 1 and ξ R. All unctions are smooth unctions o their arguments. About this system we assume the ollowing. The coeicient b(z, ξ) satisies b(z, ξ) b 0 > 0 or some b 0. Moreover, there is a positive deinite and proper smooth real-valued unction V (z) that satisies V z 0(z, 0) α( z ) z R n 1 10

11 in which α( z ) is a class K unction. This is equivalent to assume that the equilibrium z = 0 o ż = 0 (z, 0) is globally asymptotically stable. Consider now or (12) a control law u = ky in which k > 0, which yields the closed loop system ż = 0 (z, ξ) ξ = q(z, ξ) b(z, ξ)kξ. (13) For convenience, we set and rewrite the latter as in which x = ( ) z ξ ẋ = F k (x) (14) ( ) F k (x) = 0 (z, ξ). q(z, ξ) b(z, ξ)kξ Consider, or this system, the candidate Lyapunov unction W (x) = V (z) ξ2 which is positive deinite and proper. For any real number a 0, let Ω a = {x R n : W (x) a} denote the sublevel set consisting o all points o R n at which the value o W (x) is less than or equal to a and let B a = {x R n : x a} denotes the closed ball consisting o all points o R n whose norm does not exceed a. Since W (x) is proper, the set Ω a is a compact set or any a. Pick any arbitrarily large number R, any arbitrarily small number r. Since W (x) is positive deinite and proper, there exist numbers numbers 0 < d < c such that Ω d B r B R Ω c. Consider also the compact annular region S c d = {x Rn : d V (x) c}. Our goal is to show that i the gain coeicient k is large enough the unction Ẇ (x) := W x F k(x) 11

12 is negative at each point o Sd c. Observe, in this respect, that Ẇ (x) = V x 0(z, ξ) + ξq(z, ξ) b(z, ξ)kξ 2. To this end, we proceed as ollows. Consider the compact set Z = S c d {x R n : ξ = 0}. At each point o Z Ẇ (x) = V x 0(z, 0) α( z ) Since z 0 at each point o Z, there is a number a > 0 such that Ẇ (x) a x Z. Hence, by continuity, there is an open set Z ε containing Z such Ẇ (x) a/2 x Z ε. (15) Consider now the set S = {x S c d : x Z ε}. which is a compact set, let M = max{ V x S x 0(z, ξ) + ξq(z, ξ)} m = min{b(z, ξ)ξ 2 } x S and observe that m > 0 because b(z, ξ) b 0 > 0 and ξ cannot vanish at any point o S. Thus, since k > 0, we obtain Ẇ (x) M km x S. Let k 1 be such that M k 1 m = a/2. Then, i k k 1, This, together with (15) shows that Ẇ (x) a/2 x S. (16) k k 1 Ẇ (x) a/2 x S c d, as requested. This being the case, we see that or any trajectory with initial condition in Sd c, so long as z(t) Sd c we have W (x(t)) W (x(0)) (a/2)t. As a consequence, the trajectory in inite time enters the set Ω d, and remains there (because Ẇ (x) is negative on the boundary o this set). We summarize the result as ollows. 12

13 Proposition 3 Consider system (12). Suppose that b(z, ξ) b 0 > 0 or some b 0 and suppose that the equilibrium z = 0 o ż = 0 (z, 0) is globally asymptotically stable. Then, or every choice o two positive numbers R, r, with R r, there is a number k 1 such that, or all k k 1 there is a inite time T such that all trajectories o (13) with initial condition x(0) R remain bounded and satisy x(t) r or all t T. This property is commonly known as property o semi-global practical stabilizability, o the equilibrium (z, ξ) = (0, 0) o (12). To obtain asymptotic stability, an extra assumption is required. The simplest case in which this can be achieved is when the equilibrium z = 0 o ż = 0 (z, 0) is not just asymptotically, but also locally exponentially, stable, i.e. when the eigenvalues o the matrix F 0 := 0 (0, 0) (17) z have negative real part. Linear arguments can be invoked to prove the existence o a number k 2 such that, i k k 2, the equilibrium x = 0 o (12) is locally asymptotically stable. It is also possible to show that there is a number r such that, or all k k 2, the closed ball B r is always contained in the domain o attraction o x = 0. To check that this is the case, let P be the positive deinite solution o P F 0 + F T 0 P = 2I, which exists because all eigenvalues o F 0 have negative real part and consider, or the system the candidate quadratic Lyapunov unction yielding U(x) = z T P z ξ2 U(x) = 2z T P 0 (z, ξ) + ξq(z, ξ) b(z, ξ)kξ 2. Expand 0 (z, ξ) as in which Then, there is a number δ such that, and this yields 0 (z, ξ) = F 0 z + g(z) + [ 0 (z, ξ) 0 (z, 0)] g(z) lim = 0. z 0 z z δ g(z) 1 2 P z 2z T P [F 0 z + g(z)] = 2 z 2 + 2z T P g(z) z 2 z B δ. Since the unction [ 0 (z, ξ) 0 (z, 0)] is a continuously dierentiable unction that vanish at ξ = 0, there is a number M 1 such that 2z T P [ 0 (z, ξ) 0 (z, 0)] M 1 z ξ or all (z, ξ) such that z B δ and ξ δ. 13

14 Likewise, since q(z, ξ) is a continuously dierentiable unction that vanish at (z, ξ) = (0, 0), there are numbers N 1 and N 2 such that ξq(z, ξ) N 1 z ξ + N 2 ξ 2 or all (z, ξ) such that z B δ and ξ δ. Finally, since b(z, ξ) is positive and nowhere zero, there is a number b 0 such that (recall that k > 0) kb(z, ξ) kb 0 or all (z, ξ) such that z B δ and ξ δ. Putting all these inequalities together, one inds that, or all or all (z, ξ) such that z B δ and ξ δ U(x) z 2 + (M 1 + N 1 ) z ξ (kb 0 N 2 ) ξ 2. It is easy to check that i k is such that (2kb 0 2N 2 1) (M 1 + N 1 ) 2, U(x) 1 2 x 2. This shows that there is a number k 2 such that, i k k 2, the unction U(x) is negative deinite or all x satisying z B δ and ξ δ. Pick now any (nontrivial) sublevel set Ω c o U(x) entirely contained in the set o all x satisying z B δ and ξ δ and let r be such that B r Ω c. Then, the argument above shows that, or all k k 2, the equilibrium x = 0 is asymptotically stable with a domain o attraction that contains B r, which is a set independent o the choice o k. Pick now r r, pick any R > 0 and use the result o the Proposition above. There is a number k 1 such that, i k k 1 all trajectories with initial condition in B R in inite time enter the region B r, and hence enter the region o attraction o x = 0. It can be concluded, setting k = max{k 1, k 2 }, that or all k k the equilibrium x = 0 o (13) is asymptotically stable, with a domain o attraction that contains B R. The case o higher relative degree ż = 0 (z, ξ 1 ) ξ 1 = ξ 2 ξ r 1 = ξ r ξ r = q(z, ξ 1,..., ξ r ) + b(z, ξ 1,..., ξ r )u can be reduced to the case already studied. To this end, it suices to consider the dummy output variable θ = ξ r + g r 1 a 0 ξ 1 + g r 2 a 1 ξ 2 + ga r 2 ξ r 1 with a 0,..., a r 2 such that the polynomial d(λ) = λ r 1 + a r 2 λ r a 1 λ + a 0 is Hurwitz. The system thus obtained has relative degree 1 and a normal orm z = 0 ( z, θ) θ = q( z, θ) + b( z, θ)u 14

15 in which z = col(z, ξ 1,..., ξ r 1 ) 0 (z, ξ 1 ) ξ 2 0 ( z, θ) = ξ r 2 [g r 1 a 0 ξ 1 + g r 2 a 1 ξ 2 + ga r 2 ξ r 1 ] + θ and q( z, θ), b( z, θ) suitable unctions. The zero dynamics o this system are characterized by ż = 0 (z, ξ 1 ) ξ 1 = ξ 2 ξ r 1 = [g r 1 a 0 ξ 1 + g r 2 a 1 ξ 2 + ga r 2 ξ r 1 ] Changing variables as yields a system o the orm ζ i = 1 g i 1 ξ i, i = 1, 2,..., i 1, ż = 0 (z, ζ 1 ) ζ = gaζ in which A = a 0 a 1 a 2 a r 2 is a Hurwitz matrix. Arguments similar to those used above can be invoked to show that i the equilibrium z = 0 o ż = 0 (z, 0) is globally asymptotically and locally exponentially stable the equilibrium z = 0 o z = 0 ( z, 0) (18) is locally exponentially stable, with a domain o attraction that can be made arbitrarily large by increasing the design parameter g. In act, letting Q to be the positive deinite solution o QA + A T Q = I and considering the candidate Lyapunov unction W ( z) = V (z) + ζ T Qζ one proves, in exactly the same way, that or every choice o two positive numbers R, r, with R r, there is a number g 1 such that, or all g g 1 there is a inite time T such that all 15

16 trajectories o (18) with initial condition z(0) R remain bounded and satisy (t) r or all t T. Similarly, using instead the Lyapunov unction U( z) = z T P z + ζ T Qζ, one proves, in exactly the same way, that there exists a number g 2 and a number r such that, or all g g 2, the equilibrium z = 0 o (18) is locally asymptotically stable, with a domain o attraction that contains B r. In this way, the problem is reduced to the problem considered at the beginning, namely the problem o controlling a system having relative degree 1. Thus a eedback law o the orm u = kθ, with large k, yields asymptotic stability o the equilibrium ( z, θ) = (0, 0), with a domain o attraction which include a ball o arbitrarily ixed radius R. Remark. Note that, in the construction thus described, it is essential that 0 (zξ 1 ) only depends on ξ 1 and not on the other components ξ 2,..., ξ r. In the actual coordinates, the eedback law thus ound is a linear law in the ξ i s, namely u = k[ξ r + g r 1 a 0 ξ 1 + g r 2 a 1 ξ 2 + ga r 2 ξ r 1. This sometimes called a partial state eedback. Finally we observe that the result discussed so ar can be extended to the case in which the unctions characterizing the system also depend, in a smooth ashion, on a vector µ o uncertain parameters, so long as the latter ranges on a compact set. The proo proceeds essentially in the same way. 16

Feedback Linearization

Feedback Linearization Feedback Linearization Peter Al Hokayem and Eduardo Gallestey May 14, 2015 1 Introduction Consider a class o single-input-single-output (SISO) nonlinear systems o the orm ẋ = (x) + g(x)u (1) y = h(x) (2)

More information

Feedback Linearization Lectures delivered at IIT-Kanpur, TEQIP program, September 2016.

Feedback Linearization Lectures delivered at IIT-Kanpur, TEQIP program, September 2016. Feedback Linearization Lectures delivered at IIT-Kanpur, TEQIP program, September 216 Ravi N Banavar banavar@iitbacin September 24, 216 These notes are based on my readings o the two books Nonlinear Control

More information

1 The Observability Canonical Form

1 The Observability Canonical Form NONLINEAR OBSERVERS AND SEPARATION PRINCIPLE 1 The Observability Canonical Form In this Chapter we discuss the design of observers for nonlinear systems modelled by equations of the form ẋ = f(x, u) (1)

More information

Lecture : Feedback Linearization

Lecture : Feedback Linearization ecture : Feedbac inearization Niola Misovic, dipl ing and Pro Zoran Vuic June 29 Summary: This document ollows the lectures on eedbac linearization tought at the University o Zagreb, Faculty o Electrical

More information

Robust feedback linearization

Robust feedback linearization Robust eedback linearization Hervé Guillard Henri Bourlès Laboratoire d Automatique des Arts et Métiers CNAM/ENSAM 21 rue Pinel 75013 Paris France {herveguillardhenribourles}@parisensamr Keywords: Nonlinear

More information

Nonlinear Control Systems

Nonlinear Control Systems Nonlinear Control Systems António Pedro Aguiar pedro@isr.ist.utl.pt 7. Feedback Linearization IST-DEEC PhD Course http://users.isr.ist.utl.pt/%7epedro/ncs1/ 1 1 Feedback Linearization Given a nonlinear

More information

Topic # /31 Feedback Control Systems. Analysis of Nonlinear Systems Lyapunov Stability Analysis

Topic # /31 Feedback Control Systems. Analysis of Nonlinear Systems Lyapunov Stability Analysis Topic # 16.30/31 Feedback Control Systems Analysis of Nonlinear Systems Lyapunov Stability Analysis Fall 010 16.30/31 Lyapunov Stability Analysis Very general method to prove (or disprove) stability of

More information

Stabilization and Passivity-Based Control

Stabilization and Passivity-Based Control DISC Systems and Control Theory of Nonlinear Systems, 2010 1 Stabilization and Passivity-Based Control Lecture 8 Nonlinear Dynamical Control Systems, Chapter 10, plus handout from R. Sepulchre, Constructive

More information

Introduction to Nonlinear Control Lecture # 3 Time-Varying and Perturbed Systems

Introduction to Nonlinear Control Lecture # 3 Time-Varying and Perturbed Systems p. 1/5 Introduction to Nonlinear Control Lecture # 3 Time-Varying and Perturbed Systems p. 2/5 Time-varying Systems ẋ = f(t, x) f(t, x) is piecewise continuous in t and locally Lipschitz in x for all t

More information

Stability of Feedback Solutions for Infinite Horizon Noncooperative Differential Games

Stability of Feedback Solutions for Infinite Horizon Noncooperative Differential Games Stability of Feedback Solutions for Infinite Horizon Noncooperative Differential Games Alberto Bressan ) and Khai T. Nguyen ) *) Department of Mathematics, Penn State University **) Department of Mathematics,

More information

Strong Lyapunov Functions for Systems Satisfying the Conditions of La Salle

Strong Lyapunov Functions for Systems Satisfying the Conditions of La Salle 06 IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. 49, NO. 6, JUNE 004 Strong Lyapunov Functions or Systems Satisying the Conditions o La Salle Frédéric Mazenc and Dragan Ne sić Abstract We present a construction

More information

MANEUVER REGULATION, TRANSVERSE FEEDBACK LINEARIZATION, AND ZERO DYNAMICS

MANEUVER REGULATION, TRANSVERSE FEEDBACK LINEARIZATION, AND ZERO DYNAMICS MANEUVER REGULATION, TRANSVERSE FEEDBACK LINEARIZATION, AND ZERO DYNAMICS Chris Nielsen,1 Manredi Maggiore,1 Department o Electrical and Computer Engineering, University o Toronto, Toronto, ON, Canada

More information

Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems. p. 1/1

Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems. p. 1/1 Nonlinear Systems and Control Lecture # 12 Converse Lyapunov Functions & Time Varying Systems p. 1/1 p. 2/1 Converse Lyapunov Theorem Exponential Stability Let x = 0 be an exponentially stable equilibrium

More information

Output Feedback and State Feedback. EL2620 Nonlinear Control. Nonlinear Observers. Nonlinear Controllers. ẋ = f(x,u), y = h(x)

Output Feedback and State Feedback. EL2620 Nonlinear Control. Nonlinear Observers. Nonlinear Controllers. ẋ = f(x,u), y = h(x) Output Feedback and State Feedback EL2620 Nonlinear Control Lecture 10 Exact feedback linearization Input-output linearization Lyapunov-based control design methods ẋ = f(x,u) y = h(x) Output feedback:

More information

A nemlineáris rendszer- és irányításelmélet alapjai Relative degree and Zero dynamics (Lie-derivatives)

A nemlineáris rendszer- és irányításelmélet alapjai Relative degree and Zero dynamics (Lie-derivatives) A nemlineáris rendszer- és irányításelmélet alapjai Relative degree and Zero dynamics (Lie-derivatives) Hangos Katalin BME Analízis Tanszék Rendszer- és Irányításelméleti Kutató Laboratórium MTA Számítástechnikai

More information

Math 754 Chapter III: Fiber bundles. Classifying spaces. Applications

Math 754 Chapter III: Fiber bundles. Classifying spaces. Applications Math 754 Chapter III: Fiber bundles. Classiying spaces. Applications Laurențiu Maxim Department o Mathematics University o Wisconsin maxim@math.wisc.edu April 18, 2018 Contents 1 Fiber bundles 2 2 Principle

More information

Nonlinear Control. Nonlinear Control Lecture # 3 Stability of Equilibrium Points

Nonlinear Control. Nonlinear Control Lecture # 3 Stability of Equilibrium Points Nonlinear Control Lecture # 3 Stability of Equilibrium Points The Invariance Principle Definitions Let x(t) be a solution of ẋ = f(x) A point p is a positive limit point of x(t) if there is a sequence

More information

On Convexity of Reachable Sets for Nonlinear Control Systems

On Convexity of Reachable Sets for Nonlinear Control Systems Proceedings o the European Control Conerence 27 Kos, Greece, July 2-5, 27 WeC5.2 On Convexity o Reachable Sets or Nonlinear Control Systems Vadim Azhmyakov, Dietrich Flockerzi and Jörg Raisch Abstract

More information

An introduction to Mathematical Theory of Control

An introduction to Mathematical Theory of Control An introduction to Mathematical Theory of Control Vasile Staicu University of Aveiro UNICA, May 2018 Vasile Staicu (University of Aveiro) An introduction to Mathematical Theory of Control UNICA, May 2018

More information

1. Find the solution of the following uncontrolled linear system. 2 α 1 1

1. Find the solution of the following uncontrolled linear system. 2 α 1 1 Appendix B Revision Problems 1. Find the solution of the following uncontrolled linear system 0 1 1 ẋ = x, x(0) =. 2 3 1 Class test, August 1998 2. Given the linear system described by 2 α 1 1 ẋ = x +

More information

Convergence Rate of Nonlinear Switched Systems

Convergence Rate of Nonlinear Switched Systems Convergence Rate of Nonlinear Switched Systems Philippe JOUAN and Saïd NACIRI arxiv:1511.01737v1 [math.oc] 5 Nov 2015 January 23, 2018 Abstract This paper is concerned with the convergence rate of the

More information

Passivity-based Stabilization of Non-Compact Sets

Passivity-based Stabilization of Non-Compact Sets Passivity-based Stabilization of Non-Compact Sets Mohamed I. El-Hawwary and Manfredi Maggiore Abstract We investigate the stabilization of closed sets for passive nonlinear systems which are contained

More information

EN Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015

EN Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015 EN530.678 Nonlinear Control and Planning in Robotics Lecture 3: Stability February 4, 2015 Prof: Marin Kobilarov 0.1 Model prerequisites Consider ẋ = f(t, x). We will make the following basic assumptions

More information

Observability. Dynamic Systems. Lecture 2 Observability. Observability, continuous time: Observability, discrete time: = h (2) (x, u, u)

Observability. Dynamic Systems. Lecture 2 Observability. Observability, continuous time: Observability, discrete time: = h (2) (x, u, u) Observability Dynamic Systems Lecture 2 Observability Continuous time model: Discrete time model: ẋ(t) = f (x(t), u(t)), y(t) = h(x(t), u(t)) x(t + 1) = f (x(t), u(t)), y(t) = h(x(t)) Reglerteknik, ISY,

More information

Modeling and Analysis of Dynamic Systems

Modeling and Analysis of Dynamic Systems Modeling and Analysis of Dynamic Systems Dr. Guillaume Ducard Fall 2017 Institute for Dynamic Systems and Control ETH Zurich, Switzerland G. Ducard c 1 / 57 Outline 1 Lecture 13: Linear System - Stability

More information

Separation Theorems for a Class of Retarded Nonlinear Systems

Separation Theorems for a Class of Retarded Nonlinear Systems Separation Theorems or a Class o Retarded Nonlinear Systems A. Germani C. Manes P. Pepe Dipartimento di Ingegneria Elettrica e dell Inormazione 674 Poggio di Roio L Aquila Italy e-mail: alredo.germani@univaq.it

More information

Telescoping Decomposition Method for Solving First Order Nonlinear Differential Equations

Telescoping Decomposition Method for Solving First Order Nonlinear Differential Equations Telescoping Decomposition Method or Solving First Order Nonlinear Dierential Equations 1 Mohammed Al-Reai 2 Maysem Abu-Dalu 3 Ahmed Al-Rawashdeh Abstract The Telescoping Decomposition Method TDM is a new

More information

SEPARATED AND PROPER MORPHISMS

SEPARATED AND PROPER MORPHISMS SEPARATED AND PROPER MORPHISMS BRIAN OSSERMAN Last quarter, we introduced the closed diagonal condition or a prevariety to be a prevariety, and the universally closed condition or a variety to be complete.

More information

1 Lyapunov theory of stability

1 Lyapunov theory of stability M.Kawski, APM 581 Diff Equns Intro to Lyapunov theory. November 15, 29 1 1 Lyapunov theory of stability Introduction. Lyapunov s second (or direct) method provides tools for studying (asymptotic) stability

More information

Introduction to Geometric Control

Introduction to Geometric Control Introduction to Geometric Control Relative Degree Consider the square (no of inputs = no of outputs = p) affine control system ẋ = f(x) + g(x)u = f(x) + [ g (x),, g p (x) ] u () h (x) y = h(x) = (2) h

More information

2. ETA EVALUATIONS USING WEBER FUNCTIONS. Introduction

2. ETA EVALUATIONS USING WEBER FUNCTIONS. Introduction . ETA EVALUATIONS USING WEBER FUNCTIONS Introduction So ar we have seen some o the methods or providing eta evaluations that appear in the literature and we have seen some o the interesting properties

More information

Lyapunov Stability Theory

Lyapunov Stability Theory Lyapunov Stability Theory Peter Al Hokayem and Eduardo Gallestey March 16, 2015 1 Introduction In this lecture we consider the stability of equilibrium points of autonomous nonlinear systems, both in continuous

More information

Physics 5153 Classical Mechanics. Solution by Quadrature-1

Physics 5153 Classical Mechanics. Solution by Quadrature-1 October 14, 003 11:47:49 1 Introduction Physics 5153 Classical Mechanics Solution by Quadrature In the previous lectures, we have reduced the number o eective degrees o reedom that are needed to solve

More information

SEPARATED AND PROPER MORPHISMS

SEPARATED AND PROPER MORPHISMS SEPARATED AND PROPER MORPHISMS BRIAN OSSERMAN The notions o separatedness and properness are the algebraic geometry analogues o the Hausdor condition and compactness in topology. For varieties over the

More information

Nonlinear Control. Nonlinear Control Lecture # 8 Time Varying and Perturbed Systems

Nonlinear Control. Nonlinear Control Lecture # 8 Time Varying and Perturbed Systems Nonlinear Control Lecture # 8 Time Varying and Perturbed Systems Time-varying Systems ẋ = f(t,x) f(t,x) is piecewise continuous in t and locally Lipschitz in x for all t 0 and all x D, (0 D). The origin

More information

Chapter III. Stability of Linear Systems

Chapter III. Stability of Linear Systems 1 Chapter III Stability of Linear Systems 1. Stability and state transition matrix 2. Time-varying (non-autonomous) systems 3. Time-invariant systems 1 STABILITY AND STATE TRANSITION MATRIX 2 In this chapter,

More information

Feedback linearization control of systems with singularities: a ball-beam revisit

Feedback linearization control of systems with singularities: a ball-beam revisit Chapter 1 Feedback linearization control o systems with singularities: a ball-beam revisit Fu Zhang The Mathworks, Inc zhang@mathworks.com Benito Fernndez-Rodriguez Department o Mechanical Engineering,

More information

Sliding Mode Control and Feedback Linearization for Non-regular Systems

Sliding Mode Control and Feedback Linearization for Non-regular Systems Sliding Mode Control and Feedback Linearization or Non-regular Systems Fu Zhang Benito R. Fernández Pieter J. Mosterman Timothy Josserand The MathWorks, Inc. Natick, MA, 01760 University o Texas at Austin,

More information

VALUATIVE CRITERIA FOR SEPARATED AND PROPER MORPHISMS

VALUATIVE CRITERIA FOR SEPARATED AND PROPER MORPHISMS VALUATIVE CRITERIA FOR SEPARATED AND PROPER MORPHISMS BRIAN OSSERMAN Recall that or prevarieties, we had criteria or being a variety or or being complete in terms o existence and uniqueness o limits, where

More information

VALUATIVE CRITERIA BRIAN OSSERMAN

VALUATIVE CRITERIA BRIAN OSSERMAN VALUATIVE CRITERIA BRIAN OSSERMAN Intuitively, one can think o separatedness as (a relative version o) uniqueness o limits, and properness as (a relative version o) existence o (unique) limits. It is not

More information

LMI Methods in Optimal and Robust Control

LMI Methods in Optimal and Robust Control LMI Methods in Optimal and Robust Control Matthew M. Peet Arizona State University Lecture 15: Nonlinear Systems and Lyapunov Functions Overview Our next goal is to extend LMI s and optimization to nonlinear

More information

An introduction to Birkhoff normal form

An introduction to Birkhoff normal form An introduction to Birkhoff normal form Dario Bambusi Dipartimento di Matematica, Universitá di Milano via Saldini 50, 0133 Milano (Italy) 19.11.14 1 Introduction The aim of this note is to present an

More information

STAT 801: Mathematical Statistics. Hypothesis Testing

STAT 801: Mathematical Statistics. Hypothesis Testing STAT 801: Mathematical Statistics Hypothesis Testing Hypothesis testing: a statistical problem where you must choose, on the basis o data X, between two alternatives. We ormalize this as the problem o

More information

arxiv: v1 [math.gt] 20 Feb 2008

arxiv: v1 [math.gt] 20 Feb 2008 EQUIVALENCE OF REAL MILNOR S FIBRATIONS FOR QUASI HOMOGENEOUS SINGULARITIES arxiv:0802.2746v1 [math.gt] 20 Feb 2008 ARAÚJO DOS SANTOS, R. Abstract. We are going to use the Euler s vector ields in order

More information

INTRODUCTION TO REAL ANALYTIC GEOMETRY

INTRODUCTION TO REAL ANALYTIC GEOMETRY INTRODUCTION TO REAL ANALYTIC GEOMETRY KRZYSZTOF KURDYKA 1. Analytic functions in several variables 1.1. Summable families. Let (E, ) be a normed space over the field R or C, dim E

More information

Numerical Solution of Ordinary Differential Equations in Fluctuationlessness Theorem Perspective

Numerical Solution of Ordinary Differential Equations in Fluctuationlessness Theorem Perspective Numerical Solution o Ordinary Dierential Equations in Fluctuationlessness Theorem Perspective NEJLA ALTAY Bahçeşehir University Faculty o Arts and Sciences Beşiktaş, İstanbul TÜRKİYE TURKEY METİN DEMİRALP

More information

Introduction to Nonlinear Control Lecture # 4 Passivity

Introduction to Nonlinear Control Lecture # 4 Passivity p. 1/6 Introduction to Nonlinear Control Lecture # 4 Passivity È p. 2/6 Memoryless Functions ¹ y È Ý Ù È È È È u (b) µ power inflow = uy Resistor is passive if uy 0 p. 3/6 y y y u u u (a) (b) (c) Passive

More information

High-Gain Observers in Nonlinear Feedback Control. Lecture # 2 Separation Principle

High-Gain Observers in Nonlinear Feedback Control. Lecture # 2 Separation Principle High-Gain Observers in Nonlinear Feedback Control Lecture # 2 Separation Principle High-Gain ObserversinNonlinear Feedback ControlLecture # 2Separation Principle p. 1/4 The Class of Systems ẋ = Ax + Bφ(x,

More information

Zeros and zero dynamics

Zeros and zero dynamics CHAPTER 4 Zeros and zero dynamics 41 Zero dynamics for SISO systems Consider a linear system defined by a strictly proper scalar transfer function that does not have any common zero and pole: g(s) =α p(s)

More information

Converse Lyapunov theorem and Input-to-State Stability

Converse Lyapunov theorem and Input-to-State Stability Converse Lyapunov theorem and Input-to-State Stability April 6, 2014 1 Converse Lyapunov theorem In the previous lecture, we have discussed few examples of nonlinear control systems and stability concepts

More information

Balanced realization and model order reduction for nonlinear systems based on singular value analysis

Balanced realization and model order reduction for nonlinear systems based on singular value analysis Balanced realization and model order reduction for nonlinear systems based on singular value analysis Kenji Fujimoto a, and Jacquelien M. A. Scherpen b a Department of Mechanical Science and Engineering

More information

Chap. 1. Some Differential Geometric Tools

Chap. 1. Some Differential Geometric Tools Chap. 1. Some Differential Geometric Tools 1. Manifold, Diffeomorphism 1.1. The Implicit Function Theorem ϕ : U R n R n p (0 p < n), of class C k (k 1) x 0 U such that ϕ(x 0 ) = 0 rank Dϕ(x) = n p x U

More information

Math 216A. A gluing construction of Proj(S)

Math 216A. A gluing construction of Proj(S) Math 216A. A gluing construction o Proj(S) 1. Some basic deinitions Let S = n 0 S n be an N-graded ring (we ollows French terminology here, even though outside o France it is commonly accepted that N does

More information

SIO 211B, Rudnick. We start with a definition of the Fourier transform! ĝ f of a time series! ( )

SIO 211B, Rudnick. We start with a definition of the Fourier transform! ĝ f of a time series! ( ) SIO B, Rudnick! XVIII.Wavelets The goal o a wavelet transorm is a description o a time series that is both requency and time selective. The wavelet transorm can be contrasted with the well-known and very

More information

CHOW S LEMMA. Matthew Emerton

CHOW S LEMMA. Matthew Emerton CHOW LEMMA Matthew Emerton The aim o this note is to prove the ollowing orm o Chow s Lemma: uppose that : is a separated inite type morphism o Noetherian schemes. Then (or some suiciently large n) there

More information

Lecture 4. Chapter 4: Lyapunov Stability. Eugenio Schuster. Mechanical Engineering and Mechanics Lehigh University.

Lecture 4. Chapter 4: Lyapunov Stability. Eugenio Schuster. Mechanical Engineering and Mechanics Lehigh University. Lecture 4 Chapter 4: Lyapunov Stability Eugenio Schuster schuster@lehigh.edu Mechanical Engineering and Mechanics Lehigh University Lecture 4 p. 1/86 Autonomous Systems Consider the autonomous system ẋ

More information

1 Controllability and Observability

1 Controllability and Observability 1 Controllability and Observability 1.1 Linear Time-Invariant (LTI) Systems State-space: Dimensions: Notation Transfer function: ẋ = Ax+Bu, x() = x, y = Cx+Du. x R n, u R m, y R p. Note that H(s) is always

More information

Continued fractions for complex numbers and values of binary quadratic forms

Continued fractions for complex numbers and values of binary quadratic forms arxiv:110.3754v1 [math.nt] 18 Feb 011 Continued fractions for complex numbers and values of binary quadratic forms S.G. Dani and Arnaldo Nogueira February 1, 011 Abstract We describe various properties

More information

ACM/CMS 107 Linear Analysis & Applications Fall 2016 Assignment 4: Linear ODEs and Control Theory Due: 5th December 2016

ACM/CMS 107 Linear Analysis & Applications Fall 2016 Assignment 4: Linear ODEs and Control Theory Due: 5th December 2016 ACM/CMS 17 Linear Analysis & Applications Fall 216 Assignment 4: Linear ODEs and Control Theory Due: 5th December 216 Introduction Systems of ordinary differential equations (ODEs) can be used to describe

More information

Georgia Institute of Technology Nonlinear Controls Theory Primer ME 6402

Georgia Institute of Technology Nonlinear Controls Theory Primer ME 6402 Georgia Institute of Technology Nonlinear Controls Theory Primer ME 640 Ajeya Karajgikar April 6, 011 Definition Stability (Lyapunov): The equilibrium state x = 0 is said to be stable if, for any R > 0,

More information

CLASS NOTES MATH 527 (SPRING 2011) WEEK 6

CLASS NOTES MATH 527 (SPRING 2011) WEEK 6 CLASS NOTES MATH 527 (SPRING 2011) WEEK 6 BERTRAND GUILLOU 1. Mon, Feb. 21 Note that since we have C() = X A C (A) and the inclusion A C (A) at time 0 is a coibration, it ollows that the pushout map i

More information

Solution of Additional Exercises for Chapter 4

Solution of Additional Exercises for Chapter 4 1 1. (1) Try V (x) = 1 (x 1 + x ). Solution of Additional Exercises for Chapter 4 V (x) = x 1 ( x 1 + x ) x = x 1 x + x 1 x In the neighborhood of the origin, the term (x 1 + x ) dominates. Hence, the

More information

Control, Stabilization and Numerics for Partial Differential Equations

Control, Stabilization and Numerics for Partial Differential Equations Paris-Sud, Orsay, December 06 Control, Stabilization and Numerics for Partial Differential Equations Enrique Zuazua Universidad Autónoma 28049 Madrid, Spain enrique.zuazua@uam.es http://www.uam.es/enrique.zuazua

More information

Disturbance Decoupling Problem

Disturbance Decoupling Problem DISC Systems and Control Theory of Nonlinear Systems, 21 1 Disturbance Decoupling Problem Lecture 4 Nonlinear Dynamical Control Systems, Chapter 7 The disturbance decoupling problem is a typical example

More information

( x) f = where P and Q are polynomials.

( x) f = where P and Q are polynomials. 9.8 Graphing Rational Functions Lets begin with a deinition. Deinition: Rational Function A rational unction is a unction o the orm ( ) ( ) ( ) P where P and Q are polynomials. Q An eample o a simple rational

More information

Asymptotic Stability by Linearization

Asymptotic Stability by Linearization Dynamical Systems Prof. J. Rauch Asymptotic Stability by Linearization Summary. Sufficient and nearly sharp sufficient conditions for asymptotic stability of equiiibria of differential equations, fixed

More information

INVERSION IN INDIRECT OPTIMAL CONTROL

INVERSION IN INDIRECT OPTIMAL CONTROL INVERSION IN INDIRECT OPTIMAL CONTROL François Chaplais, Nicolas Petit Centre Automatique et Systèmes, École Nationale Supérieure des Mines de Paris, 35, rue Saint-Honoré 7735 Fontainebleau Cedex, France,

More information

Solutions. Problems of Chapter 1

Solutions. Problems of Chapter 1 Solutions Problems of Chapter 1 1.1 A real square matrix A IR n n is invertible if and only if its determinant is non zero. The determinant of A is a polynomial in the entries a ij of A, whose set of zeros

More information

High-Gain Observers in Nonlinear Feedback Control. Lecture # 3 Regulation

High-Gain Observers in Nonlinear Feedback Control. Lecture # 3 Regulation High-Gain Observers in Nonlinear Feedback Control Lecture # 3 Regulation High-Gain ObserversinNonlinear Feedback ControlLecture # 3Regulation p. 1/5 Internal Model Principle d r Servo- Stabilizing u y

More information

Chap. 3. Controlled Systems, Controllability

Chap. 3. Controlled Systems, Controllability Chap. 3. Controlled Systems, Controllability 1. Controllability of Linear Systems 1.1. Kalman s Criterion Consider the linear system ẋ = Ax + Bu where x R n : state vector and u R m : input vector. A :

More information

Received: 30 July 2017; Accepted: 29 September 2017; Published: 8 October 2017

Received: 30 July 2017; Accepted: 29 September 2017; Published: 8 October 2017 mathematics Article Least-Squares Solution o Linear Dierential Equations Daniele Mortari ID Aerospace Engineering, Texas A&M University, College Station, TX 77843, USA; mortari@tamu.edu; Tel.: +1-979-845-734

More information

NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION

NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION NONLINEAR CONTROL OF POWER NETWORK MODELS USING FEEDBACK LINEARIZATION Steven Ball Science Applications International Corporation Columbia, MD email: sball@nmtedu Steve Schaer Department o Mathematics

More information

Basic mathematics of economic models. 3. Maximization

Basic mathematics of economic models. 3. Maximization John Riley 1 January 16 Basic mathematics o economic models 3 Maimization 31 Single variable maimization 1 3 Multi variable maimization 6 33 Concave unctions 9 34 Maimization with non-negativity constraints

More information

Definition: Let f(x) be a function of one variable with continuous derivatives of all orders at a the point x 0, then the series.

Definition: Let f(x) be a function of one variable with continuous derivatives of all orders at a the point x 0, then the series. 2.4 Local properties o unctions o several variables In this section we will learn how to address three kinds o problems which are o great importance in the ield o applied mathematics: how to obtain the

More information

ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables

ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables Department o Electrical Engineering University o Arkansas ELEG 3143 Probability & Stochastic Process Ch. 4 Multiple Random Variables Dr. Jingxian Wu wuj@uark.edu OUTLINE 2 Two discrete random variables

More information

SPRING 2006 PRELIMINARY EXAMINATION SOLUTIONS

SPRING 2006 PRELIMINARY EXAMINATION SOLUTIONS SPRING 006 PRELIMINARY EXAMINATION SOLUTIONS 1A. Let G be the subgroup of the free abelian group Z 4 consisting of all integer vectors (x, y, z, w) such that x + 3y + 5z + 7w = 0. (a) Determine a linearly

More information

Feedback Linearizability of Strict Feedforward Systems

Feedback Linearizability of Strict Feedforward Systems Proceedings o the 47th IEEE Conerence on Decision and Control Cancun, Mexico, Dec 9-11, 28 WeB11 Feedback Linearizability o Strict Feedorward Systems Witold Respondek and Issa Amadou Tall Abstract For

More information

CHAPTER 1: INTRODUCTION. 1.1 Inverse Theory: What It Is and What It Does

CHAPTER 1: INTRODUCTION. 1.1 Inverse Theory: What It Is and What It Does Geosciences 567: CHAPTER (RR/GZ) CHAPTER : INTRODUCTION Inverse Theory: What It Is and What It Does Inverse theory, at least as I choose to deine it, is the ine art o estimating model parameters rom data

More information

MATH 205C: STATIONARY PHASE LEMMA

MATH 205C: STATIONARY PHASE LEMMA MATH 205C: STATIONARY PHASE LEMMA For ω, consider an integral of the form I(ω) = e iωf(x) u(x) dx, where u Cc (R n ) complex valued, with support in a compact set K, and f C (R n ) real valued. Thus, I(ω)

More information

Midterm Exam Solutions February 27, 2009

Midterm Exam Solutions February 27, 2009 Midterm Exam Solutions February 27, 2009 (24. Deine each o the ollowing statements. You may assume that everyone knows the deinition o a topological space and a linear space. (4 a. X is a compact topological

More information

SWEEP METHOD IN ANALYSIS OPTIMAL CONTROL FOR RENDEZ-VOUS PROBLEMS

SWEEP METHOD IN ANALYSIS OPTIMAL CONTROL FOR RENDEZ-VOUS PROBLEMS J. Appl. Math. & Computing Vol. 23(2007), No. 1-2, pp. 243-256 Website: http://jamc.net SWEEP METHOD IN ANALYSIS OPTIMAL CONTROL FOR RENDEZ-VOUS PROBLEMS MIHAI POPESCU Abstract. This paper deals with determining

More information

1. Nonlinear Equations. This lecture note excerpted parts from Michael Heath and Max Gunzburger. f(x) = 0

1. Nonlinear Equations. This lecture note excerpted parts from Michael Heath and Max Gunzburger. f(x) = 0 Numerical Analysis 1 1. Nonlinear Equations This lecture note excerpted parts from Michael Heath and Max Gunzburger. Given function f, we seek value x for which where f : D R n R n is nonlinear. f(x) =

More information

LECTURE 6: FIBER BUNDLES

LECTURE 6: FIBER BUNDLES LECTURE 6: FIBER BUNDLES In this section we will introduce the interesting class o ibrations given by iber bundles. Fiber bundles lay an imortant role in many geometric contexts. For examle, the Grassmaniann

More information

Solution of Linear State-space Systems

Solution of Linear State-space Systems Solution of Linear State-space Systems Homogeneous (u=0) LTV systems first Theorem (Peano-Baker series) The unique solution to x(t) = (t, )x 0 where The matrix function is given by is called the state

More information

ENSC327 Communications Systems 2: Fourier Representations. School of Engineering Science Simon Fraser University

ENSC327 Communications Systems 2: Fourier Representations. School of Engineering Science Simon Fraser University ENSC37 Communications Systems : Fourier Representations School o Engineering Science Simon Fraser University Outline Chap..5: Signal Classiications Fourier Transorm Dirac Delta Function Unit Impulse Fourier

More information

(x k ) sequence in F, lim x k = x x F. If F : R n R is a function, level sets and sublevel sets of F are any sets of the form (respectively);

(x k ) sequence in F, lim x k = x x F. If F : R n R is a function, level sets and sublevel sets of F are any sets of the form (respectively); STABILITY OF EQUILIBRIA AND LIAPUNOV FUNCTIONS. By topological properties in general we mean qualitative geometric properties (of subsets of R n or of functions in R n ), that is, those that don t depend

More information

MANEUVER REGULATION VIA TRANSVERSE FEEDBACK LINEARIZATION: THEORY AND EXAMPLES

MANEUVER REGULATION VIA TRANSVERSE FEEDBACK LINEARIZATION: THEORY AND EXAMPLES MANEUVER REGULATION VIA TRANSVERSE FEEDBACK LINEARIZATION: THEORY AND EXAMPLES Chris Nielsen,1 Manredi Maggiore,1 Department o Electrical and Computer Engineering, University o Toronto, Toronto, ON, Canada.

More information

Robust Semiglobal Nonlinear Output Regulation The case of systems in triangular form

Robust Semiglobal Nonlinear Output Regulation The case of systems in triangular form Robust Semiglobal Nonlinear Output Regulation The case of systems in triangular form Andrea Serrani Department of Electrical and Computer Engineering Collaborative Center for Control Sciences The Ohio

More information

Lecture 2: Controllability of nonlinear systems

Lecture 2: Controllability of nonlinear systems DISC Systems and Control Theory of Nonlinear Systems 1 Lecture 2: Controllability of nonlinear systems Nonlinear Dynamical Control Systems, Chapter 3 See www.math.rug.nl/ arjan (under teaching) for info

More information

A strongly polynomial algorithm for linear systems having a binary solution

A strongly polynomial algorithm for linear systems having a binary solution A strongly polynomial algorithm for linear systems having a binary solution Sergei Chubanov Institute of Information Systems at the University of Siegen, Germany e-mail: sergei.chubanov@uni-siegen.de 7th

More information

Handout 2: Invariant Sets and Stability

Handout 2: Invariant Sets and Stability Engineering Tripos Part IIB Nonlinear Systems and Control Module 4F2 1 Invariant Sets Handout 2: Invariant Sets and Stability Consider again the autonomous dynamical system ẋ = f(x), x() = x (1) with state

More information

Least Squares Approximate Feedback Linearization. Abstract. We study the least squares approximate feedback linearization problem:

Least Squares Approximate Feedback Linearization. Abstract. We study the least squares approximate feedback linearization problem: Least Squares Approximate Feedback Linearization Andrzej Banaszuk y Andrzej Swiech z John Hauser x Abstract. We study the least squares approximate eedback linearization problem: given a single input nonlinear

More information

Exercise Solutions to Functional Analysis

Exercise Solutions to Functional Analysis Exercise Solutions to Functional Analysis Note: References refer to M. Schechter, Principles of Functional Analysis Exersize that. Let φ,..., φ n be an orthonormal set in a Hilbert space H. Show n f n

More information

Global Weak Solution of Planetary Geostrophic Equations with Inviscid Geostrophic Balance

Global Weak Solution of Planetary Geostrophic Equations with Inviscid Geostrophic Balance Global Weak Solution o Planetary Geostrophic Equations with Inviscid Geostrophic Balance Jian-Guo Liu 1, Roger Samelson 2, Cheng Wang 3 Communicated by R. Temam) Abstract. A reormulation o the planetary

More information

A conjecture on sustained oscillations for a closed-loop heat equation

A conjecture on sustained oscillations for a closed-loop heat equation A conjecture on sustained oscillations for a closed-loop heat equation C.I. Byrnes, D.S. Gilliam Abstract The conjecture in this paper represents an initial step aimed toward understanding and shaping

More information

Theorem Let J and f be as in the previous theorem. Then for any w 0 Int(J), f(z) (z w 0 ) n+1

Theorem Let J and f be as in the previous theorem. Then for any w 0 Int(J), f(z) (z w 0 ) n+1 (w) Second, since lim z w z w z w δ. Thus, i r δ, then z w =r (w) z w = (w), there exist δ, M > 0 such that (w) z w M i dz ML({ z w = r}) = M2πr, which tends to 0 as r 0. This shows that g = 2πi(w), which

More information

Nonlinear Control Systems

Nonlinear Control Systems Nonlinear Control Systems António Pedro Aguiar pedro@isr.ist.utl.pt 3. Fundamental properties IST-DEEC PhD Course http://users.isr.ist.utl.pt/%7epedro/ncs2012/ 2012 1 Example Consider the system ẋ = f

More information

Linearization problem. The simplest example

Linearization problem. The simplest example Linear Systems Lecture 3 1 problem Consider a non-linear time-invariant system of the form ( ẋ(t f x(t u(t y(t g ( x(t u(t (1 such that x R n u R m y R p and Slide 1 A: f(xu f(xu g(xu and g(xu exist and

More information

Stability theory is a fundamental topic in mathematics and engineering, that include every

Stability theory is a fundamental topic in mathematics and engineering, that include every Stability Theory Stability theory is a fundamental topic in mathematics and engineering, that include every branches of control theory. For a control system, the least requirement is that the system is

More information

Roberto s Notes on Differential Calculus Chapter 8: Graphical analysis Section 1. Extreme points

Roberto s Notes on Differential Calculus Chapter 8: Graphical analysis Section 1. Extreme points Roberto s Notes on Dierential Calculus Chapter 8: Graphical analysis Section 1 Extreme points What you need to know already: How to solve basic algebraic and trigonometric equations. All basic techniques

More information