An observation algorithm for nonlinear systems with unknown inputs
|
|
- Erin Logan
- 5 years ago
- Views:
Transcription
1 An observaton algorthm or nonlnear systems wth unknown nputs Jean-Perre Barbot, Drss Boutat, Therry Floquet To cte ths verson: Jean-Perre Barbot, Drss Boutat, Therry Floquet An observaton algorthm or nonlnear systems wth unknown nputs Automatca, Elsever, 2009, 45 (8), pp <101016/jautomatca > <nra > HAL Id: nra Submtted on 4 Jun 2009 HAL s a mult-dscplnary open access archve or the depost and dssemnaton o scentc research documents, whether they are publshed or not The documents may come rom teachng and research nsttutons n France or abroad, or rom publc or prvate research centers L archve ouverte plurdscplnare HAL, est destnée au dépôt et à la duson de documents scentques de nveau recherche, publés ou non, émanant des établssements d ensegnement et de recherche ranças ou étrangers, des laboratores publcs ou prvés
2 An observaton algorthm or nonlnear systems wth unknown nputs Jean-Perre Barbot 1,4, Drss Boutat 2 and T Floquet 3,4 1 Equpe Commande des Systèmes (ECS), ENSEA, 6 Avenue du Ponceau, Cergy, France 2 LVR/ENSIB, 10 Boulevard de Lahtolle, Bourges, France 3 LAGIS UMR CNRS 8146, Ecole Centrale de Llle, BP 48, Cté Scentque, Vlleneuve-d Ascq, France 4 Equpe Projet ALIEN, INRIA Llle - Nord Europe, France Abstract Ths paper provdes some new developments n the desgn o unknown nput observers or nonlnear systems An algorthm whch states the state and the unknown nput o the system can be recovered n nte tme s ntroduced Ths algorthm leads to the transormaton o the system nto an extended block trangular observable orm sutable or the desgn o nte tme observers The proposed method s useul to relax some restrctve condtons o exstng nonlnear unknown nput observer desgn procedures 1 Introducton It s o mportance to desgn observers or multvarable lnear or nonlnear systems partally drven by unknown nputs Such a problem arses n systems subject to dsturbances or wth naccessble nputs and n many applcatons such as parameter dentcaton, ault detecton and solaton or cryptography The desgn o observers or nonlnear systems s a challengng problem, even or accurately known systems It has receved a consderable amount o attenton n the lterature In many approaches, nonlnear coordnate transormatons are used to transorm the system nto sutable observer canoncal orms Then, observers wth lnearzable error dynamcs (see eg [9, 20, 21]), hgh gan observers [8] or backsteppng observers [15] can be desgned Few works deal wth the desgn o unknown nput observers or nonlnear systems Some o them are concerned wth applcatons n the eld o ault detecton and dentcaton, and n partcular the nonlnear Fundamental Problem o Resdual Generaton [4, 10, 19] Other ones deal wth nonlnear systems subject to exogenous perturbatons and are based on sldng mode consderatons [22] The convergence s obtaned n nte tme under the assumptons that the system can be put 1
3 nto a set o trangular observable orms, where the unknown nputs act only on the last dynamcs o each trangular orm Ths assumpton s known as the observablty matchng condton The present work ams at the development o a systematc method leadng to the nte tme observaton o a class o nonlnear systems wth unknown nputs even the observablty matchng condton s not ullled To ths end, the procedure gven n [5] s extended to the nonlnear case It also results n a constructve algorthm that transorms the system nto a smlar type o block trangular observable orms Ths transormaton reles on the ntroducton o sutable cttous outputs Sucent condtons or the exstence o such auxlary varables are gven Ater transormaton o the system, t s shown that t s possble to desgn observers that provde the nte tme estmaton o both the state varables and the unknown nputs An llustratve example hghlghts the ecency o the proposed methodology 2 Problem statement and motvatons Consder, on an open set U, the nonlnear system: ẋ = (x) + g(x)w = (x) + m g (x)w =1 (1) y = h(x) = [h 1 (x),, h p (x)] T where x U R n s the state vector, y R p s the output vector and where w = [w 1,, w m ] R m represents the unknown nputs The vector elds and g 1,,g m, and the unctons h 1,, h p, are assumed to be sucently smooth on U Wthout loss o generalty, t s assumed that p m, and that or all x U, the dstrbuton G = span {g 1,,g m } and the codstrbuton span {dh 1,,dh p } are nonsngular on U Lke n [17], let us rst dene the unknown nput characterstc ndexes {ρ 1,, ρ p } such that, or 1 p: L gj L k h (x) = 0, or k < ρ 1, and or all 1 j m, L gj L ρ 1 h (x) 0, or at least one 1 j m, or all x U The system (1) wth w = 0 s supposed to be locally weakly observable on U [11] Thus, there exsts a change o coordnates ( φ = h 1,, ) T h 1,, h p,, L νp 1 h p L ν1 1 such that, ater a sutable reorderng o the state components, the system (1) s locally transormed nto ξ = A δ ξ + H δ V δ (x,w) (2) η = a(ξ,η) + b(ξ,η)w (3) y = C δ ξ 2
4 wth ξ = ( ) ξ1 T,, ξp T T, ξ R δ, 1 p V δ (x, w) = L δ h (x) + m L gj L δ 1 h (x)w j R = A δ = R δ δ H δ = ( ) T R δ, C δ = ( ) R 1 δ and a, b are smooth vector elds on U The ntegers (ν 1,ν 2,,ν p ) are the so-called observablty ndces o system (1) (see [14] or a denton) and thus satsy ν ν p = n From the denton o the ρ, one has δ = mn (ν,ρ ) and g δ = 0 ν < ρ (e δ = ν ) 21 A trangular observable orm Most exstng nonlnear observers or system (1) are desgned under the assumpton that the system satses the so-called observablty matchng condton Ths condton was rst ormulated or SISO lnear systems n [18], Chapter 4 In the general case o MIMO nonlnear systems, a necessary and sucent condton s gven by: ν ρ or all 1 p (4) Then, δ = ν or all 1 p, and under the change o coordnates ξ = φ(x) the system (1) s transormed nto the orm: ξ = A ν ξ + H ν V ν (ξ,w) (5) y = C ν ξ wth m L gj L ν 1 h (x)w j 0 and only ν = ρ Each subsystem o (5) s n =1 the so-called trangular observable orm Fnte tme observers or such a orm can be ound n the lterature, based on step-by-step sldng mode technques [22], hgher order sldng modes [6, 16], numercal approaches [3, 13], or algebrac methods [2] The desgn o such observers s let to the reader, snce they are straghtorward applcatons o exstng results Nevertheless, dependng on the choce o the observer, some assumptons have to be ntroduced For nstance, n all the prevously mentoned works, the unknown nput w has, at least, to be bounded A necessary and sucent condton or the recovery o the unknown nputs s that Γ(x) = L g1 L ρ1 1 h 1 (x) L gm L ρ1 1 h 1 (x) L g1 L ρp 1 h p (x) L gm L ρp 1 h p (x) 3
5 has rank m In ths case, ρ = {ρ 1,, ρ p } s the vector relatve degree as dened n [12], p 220, when m = p 22 An extended trangular observable orm The am o ths paper s to provde an observaton algorthm that allows or the nte tme estmaton o both the state and the unknown nputs o (1) even ν j > ρ j or at least a j n {1,, p} Consder agan the general orm (2-3) Applyng any o the aorementoned nte tme observers: () or 1 p, ξ can be estmated n nte tme; () one can also recover n nte tme the last component V δ o each subsystem o (2) The problem s to recover the remanng state η Denote: where V (x) = V δ1 (x,w) V δ2 (x,w) V δp (x,w) = L δ1 h 1(x) L δ2 h 2(x) L δp h p(x) + Γ δ(x)w L g1 L δ1 1 h 1 (x) L gm L δ1 1 h 1 (x) Γ δ (x) = L g1 L δp 1 h p (x) L gm L δp 1 h p (x) Let be the commutatve algebra o the measured outputs and ther successve Le dervatves up to order δ : = span{h 1,, L δ1 1 h 1,, h p,,l δp 1 h p } and let d be the codstrbuton: d = span{dh 1,,dL δ1 1 h 1,,dh p,,dl δp 1 h p } Assume there exsts a 1 p row vector K(x) = (k 1 (x),, k p (x)) 0, k or 1 p, such that: K(x)Γ δ (x) = 0 or all x U (6) and set: ȳ = h(x) = K(x)V (x) = p k (x)l δ h (x) (7) Note that ȳ s an avalable normaton (ater a nte tme) and s not aected by the unknown nputs Thereore, d +span { d h } d, ȳ can be consdered as an addtonal cttous output Then, let ρ and ν be the unknown nput characterstc ndexes and the observablty ndces o (1) wth respect to the extended output [ y T, ȳ ] T = [ h T, h(x) ] T I ν ρ or all 1 p + 1, the =1 4
6 system (1) can be transormed nto: ξ = A ν ξ + H ν V ν ( ξ, w) or 1 p y = C ν ξ ξ = A νp+1 ξ + H νp+1 V νp+1 ( ξ,w) ȳ = C νp+1 ξ where ξ = ( ξ T 1,,ξ T p, ξ T) T R n Then, t s possble to recover both the state and the unknown nputs n nte tme Let us gve sucent condtons or the exstence o a sutable cttous output ȳ For ths, the ollowng notatons are ntroduced: ) G = span{ 1,, n m }, the annhlator o G ( are 1-orms such that ι gk = 0, where ι g s the nner product o the vector eld g and the 1-orm ) ) Ω, the module spanned by d over Proposton 1 The ollowng condtons are equvalent: ) Equaton (6) has a soluton K and KV / ) Ξ = span{ G Ω such that ι / } = {0} Proo: Set = p k dl δ 1 h wth k Clearly, Ω and KΓ δ = K =1 p ι = ι k dl δ 1 h = =1 dl δ1 1 h 1 dl δp 1 h p p =1 k L δ h = KV = ȳ [ ] [ ] g1 g m = g1 g m Thus: {K s a soluton o (6) such that KV / } { Ω,ι τ = 0 or any τ G and ι / } { G Ω and ι / } Ξ {0} The dscusson above can be recursvely generalzed as ollows Assume that the condton (4) s not stll satsed wth the extended output obtaned wth the solutons o (6) On the bass o ths new output, the correspondng matrx Γ δ can be computed and another set o cttous outputs can eventually be ound One can terate ths procedure untl the condton (4) s ullled or a new extended output Then, the orgnal system can be put nto an extended 5
7 block trangular observable orm 1 : ξ 1 = A ν 1 ξ 1 + H ν 1 V ν 1 (ξ, w) y 1 = y = C ν 1 ξ 1, 1 p 1 ξ 2 = A ν 2 ξ 2 + H ν 2 V ν 2 (ξ, w) y 2 = C ν 2 ξ 2, 1 p 2 ξ k y k = A ν k ξ k + H ν k V ν k (ξ, w) = C ν k ξ k, 1 p k (8) where the ntegers ν j are the observablty ndces o the system (1) wth the new outputs y j The rst subsystem s ed by the orgnal outputs o the system A nte tme observer s desgned to estmate the state o ths subsystem and to provde n nte tme the knowledge o the cttous outputs y 2, 1 p2 Then, the state o the second trangular observable orm can be estmated as well as the cttous outputs y 3 Thus, one can recursvely obtan the whole state o the system n nte tme Remark 1 The nte tme property s requred to ensure that one obtans a ast and accurate estmaton o the cttous outputs (or nstance, va the equvalent output njecton n the case o sldng mode observers, see [6] and the reerences theren) Furthermore, ths property s oten desrable n the ramework o observaton and partcularly or the purpose o observer-based controller desgn or nonlnear systems Then, or a large class o nonlnear systems, the observer can be desgned separately rom the controller and the separaton prncple does not need to be proved It can also be o paramount mportance n applcatons that requre ast estmatons o some unknown nputs lke ault detecton and dentcaton or on-lne parameter dentcaton 3 Nonlnear unknown nput observer algorthm An algorthm that states the system can be transormed nto (8) and that provdes the ntegers p j, ν j and the auxlary outputs y j (j = 1,,k ) s now gven [ ] T Step 0: Compute G and ts annhlator G Set p 1 = p, h 1 1,, h 1 p = 1 [h 1,,h p ] T, µ 0 = 0, z 0 1 = = z 0 µ 0 = 0 Step α: [a] Consder y α = [ h α 1,,h α p α ] T R p α and reorder ts components as ollows: y α = [ h α 1,, h α l α,hα l α +1,, h α p α ] T 1 Systems that admt such a orm belong to the class o let nvertble systems wth trval zero dynamcs (see [2]) 6
8 such that or 1 j l α : [1,,m], k N L g L k h α j = 0 and or 1 j p α l α, there exsts an nteger ρ α j such that: [1,,m] L g L k h α l α +j = 0 k < ρ α j 1 [1,,m] L g L ρα j 1 h α l α +j 0 [b] Dene Φ α = {dh α 1,, dl n 1 h α 1,, dh α l I α = span α,, dln 1 {( dz α 1 1,,dz α 1 µ α 1 ) Φ α } Let dmi α = µ α 1 + ϕ α I α can be wrtten as ollows: h α lα} Compute I α = span{dz1 α 1,, dz α 1 µ,dh α α 1 1,,dL ϕα 1 1 h α 1,, dh α l α,,dlϕα l α 1 h α l α} wth lα ϕ α = ϕα I µ α 1 + ϕ α = n, set =1 {dz k 1,, dz k µ k } µ k = µ α 1 + ϕ α = {dz α 1 1,, dz α 1 µ,dh α α 1 1,, dl ϕα 1 1 h α 1,, dh α l α,, dlϕα l α 1 h α l α} and stop the algorthm [c] I µ α 1 + ϕ α < n, consder the outputs aected by the unknown nputs and dene: Υ α = {dh α l +1,, dl ρα α 1 1 h α l +1,, dh α α p α,, dlρα p α l α 1 h α p α} Compute the codstrbuton Ω α = span {I α Υ α } Let dmω α = µ α 1 + ϕ α + κ α = µ α and wrte Ω α = span{dz1 α,,dzµ α α} wth and pα l α =1 {dz1 α,,dzµ α α} = {dzα 1 1,,dz α 1 µ, dh α α 1 1,, dl ϕα 1 1 h α 1, h α l α,dhα l α +1,, dlκα 1 1 h α l α +1,, dh α l α,, dlϕα l α 1, dh α p α,,dlκα p α l α 1 h α p α} κ α = κα I µ α = n, the algorthm stops [d] Otherwse, µ α < n Dene α = span{z1 α,, zµ α { α} µ α } Ω α = span φ dz α, φ α =1 Ξ α = span { G Ω α such that ι / α} 7
9 Let p α+1 = dm Ξ α I p α+1 = 0, the state o the system (1) can not be recovered wth the method descrbed n ths paper and the algorthm stops Otherwse, there exst p α+1 one-orms such that Ξ α = span { 1,, pα+1} and one can dene the ollowng vector o cttous outputs, sutable to the problem (see Proposton 1): y α+1 = [ ] ] T T ι 1,,ι Set [h α+1 pα+1 1,, h α+1 p = [ ] α+1 T ι 1,,ι Go to [a] pα+1 I the algorthm stops or some µ k = n, the change o coordnates φ = (z k 1,,z k µ k ) T s well dened and transorms the system nto a set o block trangular observable orms smlar to (8) The unctons h j, and the ntegers p, l, ϕ j, κ j are obtaned n each -th teraton o the algorthm Remark 2 I the condton (4) s satsed or the measured outputs o the system (1), µ 1 = n ater the rst teraton o the algorthm and the system s exactly transormed nto the orm (5) that s usually consdered or the desgn o asymptotc (see eg [8, 15]) or robust nte tme ([6, 22]) nonlnear observers 4 Estmaton o the unknown nputs I the algorthm ends n a postve way, the unknown nputs can also be obtaned n a nte tme Indeed, the use o a nte tme observer provdes an estmaton o the state, say x, and the knowledge o the ollowng quanttes (rom the last lne o each block o the trangular orm): Θ( x) = m θ j = L κ j h l +j ( x) + L gs L κ j 1 h l +j ( x)w s, (9) s=1 or 1 j p l, and 1 k The relatons (9) can be rewrtten as: Λ( x)w = Θ( x) wth L g1 L κ1 1 1 h 1 l 1 +1 ( x) L g m L κ1 1 1 h 1 l 1 +1 ( x) Λ( x) = L g1 L κk p k l 1 h k k ( x) L p g m L κ k p k l 1 h k k ( x) p θ 1 1 L κ1 1 h1 l 1 +1 ( x) θ k L κ p k l k p k l h k p k ( x) 8
10 Snce the dstrbuton span {g 1,,g m } s assumed to be nonsngular, the matrx dl κ1 1 1 h 1 l 1 +1 ( x) [ ] Λ( x) = g1 g m dl κk p k l 1 h k p k ( x) has rank m on every subset o U where at least m one-orms dl κ j 1 h l +j does not belong to G Thus, an estmaton o the unknown nput s gven by: w = Λ + ( x)θ( x) where Λ + s a well dened pseudo-nverse o Λ 5 Example As a way o llustraton 2, consder the ollowng nonlnear system subject to the unknown nput w = [w 1,w 2 ] T ẋ 1 = x 2 x 3 1 ẋ 2 = x 3 + x 2 2 x a(x 3,x 4 )w 1 ẋ 3 = x 5 (10) ẋ 4 = x 4 + x b(x 2,x 3 )w 1 ẋ 5 = x 3 + x 2 w 2 ẋ 6 = x 6 + w 2 wth outputs y 1 = x 1, y 2 = x 4 and y 3 = x 6 The scalar unctons a and b are such that a(x 3, x 4 ) = a 1 (x 3 )a 2 (x 4 ), b(x 2,x 3 ) = a 1 (x 3 )b 2 (x 2 ) and where a(0,0) 0 and b(0,0) 0, and b 2 (x 2 ) 0, x 2 R For ths system, one has ρ 1 = 2, ρ 2 = ρ 3 = 1 and ν 1 = 4, ν 2 = ν 3 = 1 (as a consequence δ 1 = 2, δ 2 = δ 3 = 1) Thus, the necessary and sucent condton (4) or a system to be transormed nto a orm smlar to (5) s not ullled Furthermore, the dstrbuton G s not nvolutve As a consequence, the sldng mode observer proposed n [22], where the dstrbuton spanned by the unknown nput channels has to be nvolutve, can not be desgned here However, the procedure proposed n ths paper s applcable The annhlator o G s gven by: G = span{dx 1, dx 3, b 2 dx 2 a 2 dx 4,dx 5 x 2 dx 6 } In the rst step o the algorthm, I 1 s empty and Ω 1 = span{dh 1,dL h 1,dh 2, dh 3 } = span{dx 1, 3x 2 1dx 1 + dx 2,dx 4,dx 6 } wth dm Ω 1 = 4 < n One has 1 = span{h 1,L h 1,h 2,h 3 } = span{x 1,x 2 x 3 1,x 4,x 6 } = span{x 1,x 2,x 4,x 6 } 2 A physcal applcaton o the proposed algorthm n the eld o chaotc synchronzaton or secure communcaton can be ound n [2] 9
11 and b 2 dx 2 a 2 dx 4 Ω 1 Then G Ω 1 = span{dx 1, b 2 dx 2 a 2 dx 4 } Ξ 1 = span{b 2 dx 2 a 2 dx 4 } Ξ 1 0 and one can dene the ollowng cttous output: y 2 = ι (b 2 dx 2 a 2 dx 4 ) = b 2 (x 3 + x 2 2 x 3 2) a 2 ( x 4 + x 2 2) = b 2 (x 2 )x 3 mod[ ] Snce b 2 (x 2 ) 0 or all x 2, the knowledge o y 2 s equvalent to the knowledge o x 3 The second step starts by denng the output vector: y 2 = x 3 I 2 s empty and Ω 2 = span{dx 1,dx 2,dx 4,dx 6,dx 3,dx 5 } wth dm Ω 2 = 6 Then, one can dene the change o coordnates: z = ψ(x) = [x 1,x 2,x 4,x 6,x 3,x 5 ] T In the new coordnates, the system s rewrtten as: ż 1 = z 2 z 3 1 ż 2 = z 5 + z2 2 z2 3 + a 1 (z 5 )a 2 (z 3 )w 1 ż 3 = z 3 + z2 2 + a 1 (z 5 )b 2 (z 2 )w 1 (11) ż 4 = z 4 + w 2 ż 5 = z 6 ż 6 = z 5 + z 2 w 2 By the means o a nte tme observer ed by the orgnal outputs o system (10), e y 1 = z 1, y 2 = z 3 and y 3 = z 4, one can recover the state z 2 and V 1 = z 5 + z 2 2 z a 1 (z 5 )a 2 (z 3 )w 1 V 2 = z 3 + z a 1 (z 5 )b 2 (z 2 )w 1 V 3 = z 4 + w 2 For the sake o place, the desgn o the observer s not gven here However, the reader can or nstance reer to [7] where an example o hgher order sldng mode observer, desgned or a smlar problem n the lnear case, can be extended to the problem o nonlnear systems Then, y 2 = ι (b 2 (z 2 )dz 2 a 2 (z 3 )dz 3 ) = b 2 (z 2 ) ( z 5 + z 2 2 z a 1 (z 5 )a 2 (z 3 )w 1 ) a 2 (z 3 ) ( z a 1 (z 5 )b 2 (z 2 )w 1 ) = b2 (z 2 )V 1 a 2 (z 3 )V 2 = b 2 (z 2 )(z 5 + z 2 2 z 3 2) a 2 (z 3 )( z 3 + z 2 2) s known ater a nte tme and z 5 = y2 +a 2(z 3)( z 3+z 2 2 ) b 2(z 2) + z2 3 z2 2 s an avalable normaton Agan, a nte tme observer leads to the recovery o z 6 Then, the unknown nput can also be obtaned snce: w 1 = V1 z5 z2 2 +z3 2 a 1(z 5)a 2(z 3) and w 2 = V 3 +z 4 10
12 6 Concluson In ths paper, the problem o the observaton o nonlnear systems subject to unknown nputs was consdered An observaton algorthm that determnes whether t s possble to recover the state and unknown nput n nte tme was ntroduced When the answer s yes, the algorthm provdes a change o coordnates that transorms the system n a new type o block trangular observable orm well suted to the desgn o nte tme observers The observablty matchng condton usually requred or the desgn o nonlnear unknown nput observers s relaxed wth the proposed method Reerences [1] J-P Barbot, I Belmouhoub and L Boutat-Baddas, Observablty Normal Forms, n: New trends n Nonlnear dynamcs and control, LNCIS 295, W Kang et al, Eds, Sprnger Verlag, 2003, pp 1 24 [2] J-P Barbot, M Fless, T Floquet, An algebrac ramework or the desgn o nonlnear observers wth unknown nputs, IEEE Con on Decson and Control, New-Orleans, USA, 2007 [3] S Dop, J W Grzzle and F Chaplas, On numercal derentaton algorthms or nonlnear estmaton, n IEEE Conerence on Decson and Control, pp , 2000 [4] T Floquet, J-P Barbot, W Perruquett and M Djemaï, On the robust ault detecton va a sldng mode dsturbance observer, Internatonal Journal o control, vol 77, 2004, pp [5] T Floquet and J-P Barbot, An observablty orm or lnear systems wth unknown nputs, Internatonal Journal o control, vol 79, 2006, pp [6] T Floquet, JP Barbot, Super twstng algorthm based step-by-step sldng mode observers or nonlnear systems wth unknown nputs, Internatonal Journal o Systems Scence, vol 38, 2007, pp [7] T Floquet and JP Barbot, A canoncal orm or the desgn o unknown nput sldng mode observers, n Advances n Varable Structure and Sldng Mode Control, Lecture Notes n Control and Inormaton Scences, Vol 334, C Edwards, E Fossas Colet, L Frdman, (Eds), Sprnger Edton, 2006 [8] J P Gauther, H Hammour and S Othman, A smple observer or nonlnear systems wth applcatons to boreactors, IEEE Transactons on Automatc Control, Vol 37, 1992, pp
13 [9] A Glumneau, C H Moog, and F Plestan, New Algebro-Geometrc Condtons or the Lnearzaton by Input-Output Injecton, IEEE Transactons on Automatc Control, vol 41, 1996, pp [10] H Hammour, M Knnaert, and E H El Yaagoub, Observer-Based Approach to Fault Detecton and Isolaton or Nonlnear Systems, IEEE Transactons on Automatc Control, vol 44, 1999, pp [11] R Hermann and A J Krener, Nonlnear controllablty and observablty, IEEE Transactons on Automatc Control, Vol 22, pp , 1977 [12] A Isdor, Nonlnear Control Systems, Communcaton and Control Engneerng Seres, Thrd edton, Sprnger-Verlag, 1995 [13] W Kang, Movng Horzon Numercal Observers o Nonlnear Control Systems, IEEE Transactons on Automatc Control, Vol 51, No 2, pp , 2006 [14] A J Krener and W Respondek, Nonlnear observers wth lnearzable error dynamcs, SIAM J Control Optm, Vol 23, 1985, pp [15] A J Krener and W Kang, Locally convergent nonlnear observers, SIAM J Control Optm, Vol 42, , 2003 [16] A Levant, Robust Exact Derentaton va sldng mode technque, Automatca, Vol 34, No 3, pp , 1998 [17] R Marno, W Respondek, A J Van der Schat, Almost dsturbance decouplng or sngle-nput sngle-output nonlnear systems, IEEE Transactons on Automatc Control, Vol 34, pp , 1989 [18] W Perruquett and J-P Barbot, Sldng Mode Control n Engneerng, Ed Marcel Dekker, 2002 [19] C de Perss and A Isdor, A geometrc approach to nonlnear ault detecton and solaton, IEEE Transactons on Automatc Control, vol 46, 2001, pp [20] W Respondek, A Pogromsky, H Njmejer, Tme scalng or observer desgn wth lnearzable error dynamcs, Automatca, Vol 40, No 2, pp , 2004 [21] X H Xa and W B Gao, Nonlnear observer desgn by observer error lnearzaton, SIAM J Control Optm, vol 27, 1989, pp [22] Y Xong and M Sa, Sldng Mode Observer or Nonlnear Uncertan Systems, IEEE Transactons on Automatc Control, vol 46, 2001, pp
SINGLE OUTPUT DEPENDENT QUADRATIC OBSERVABILITY NORMAL FORM
SINGLE OUTPUT DEPENDENT QUADRATIC OBSERVABILITY NORMAL FORM G Zheng D Boutat JP Barbot INRIA Rhône-Alpes, Inovallée, 655 avenue de l Europe, Montbonnot Sant Martn, 38334 St Ismer Cedex, France LVR/ENSI,
More informationA generalization of a trace inequality for positive definite matrices
A generalzaton of a trace nequalty for postve defnte matrces Elena Veronca Belmega, Marc Jungers, Samson Lasaulce To cte ths verson: Elena Veronca Belmega, Marc Jungers, Samson Lasaulce. A generalzaton
More information, rst we solve te PDE's L ad L ad n g g (x) = ; = ; ; ; n () (x) = () Ten, we nd te uncton (x), te lnearzng eedbac and coordnates transormaton are gve
Freedom n Coordnates Transormaton or Exact Lnearzaton and ts Applcaton to Transent Beavor Improvement Kenj Fujmoto and Tosaru Suge Dvson o Appled Systems Scence, Kyoto Unversty, Uj, Kyoto, Japan suge@robotuassyoto-uacjp
More informationChapter 3 Differentiation and Integration
MEE07 Computer Modelng Technques n Engneerng Chapter Derentaton and Integraton Reerence: An Introducton to Numercal Computatons, nd edton, S. yakowtz and F. zdarovsky, Mawell/Macmllan, 990. Derentaton
More informationOPTIMISATION. Introduction Single Variable Unconstrained Optimisation Multivariable Unconstrained Optimisation Linear Programming
OPTIMIATION Introducton ngle Varable Unconstraned Optmsaton Multvarable Unconstraned Optmsaton Lnear Programmng Chapter Optmsaton /. Introducton In an engneerng analss, sometmes etremtes, ether mnmum or
More information36.1 Why is it important to be able to find roots to systems of equations? Up to this point, we have discussed how to find the solution to
ChE Lecture Notes - D. Keer, 5/9/98 Lecture 6,7,8 - Rootndng n systems o equatons (A) Theory (B) Problems (C) MATLAB Applcatons Tet: Supplementary notes rom Instructor 6. Why s t mportant to be able to
More informationChapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems
Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons
More informationMMA and GCMMA two methods for nonlinear optimization
MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons
More informationLecture 2 Solution of Nonlinear Equations ( Root Finding Problems )
Lecture Soluton o Nonlnear Equatons Root Fndng Problems Dentons Classcaton o Methods Analytcal Solutons Graphcal Methods Numercal Methods Bracketng Methods Open Methods Convergence Notatons Root Fndng
More informationInner Product. Euclidean Space. Orthonormal Basis. Orthogonal
Inner Product Defnton 1 () A Eucldean space s a fnte-dmensonal vector space over the reals R, wth an nner product,. Defnton 2 (Inner Product) An nner product, on a real vector space X s a symmetrc, blnear,
More informationModule 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:
More informationRobust observed-state feedback design. for discrete-time systems rational in the uncertainties
Robust observed-state feedback desgn for dscrete-tme systems ratonal n the uncertantes Dmtr Peaucelle Yosho Ebhara & Yohe Hosoe Semnar at Kolloquum Technsche Kybernetk, May 10, 016 Unversty of Stuttgart
More informationTransfer Functions. Convenient representation of a linear, dynamic model. A transfer function (TF) relates one input and one output: ( ) system
Transfer Functons Convenent representaton of a lnear, dynamc model. A transfer functon (TF) relates one nput and one output: x t X s y t system Y s The followng termnology s used: x y nput output forcng
More informationA Simple Research of Divisor Graphs
The 29th Workshop on Combnatoral Mathematcs and Computaton Theory A Smple Research o Dvsor Graphs Yu-png Tsao General Educaton Center Chna Unversty o Technology Tape Tawan yp-tsao@cuteedutw Tape Tawan
More informationEndogenous timing in a mixed oligopoly consisting of a single public firm and foreign competitors. Abstract
Endogenous tmng n a mxed olgopoly consstng o a sngle publc rm and oregn compettors Yuanzhu Lu Chna Economcs and Management Academy, Central Unversty o Fnance and Economcs Abstract We nvestgate endogenous
More informationContinuous Belief Functions: Focal Intervals Properties.
Contnuous Belef Functons: Focal Intervals Propertes. Jean-Marc Vannobel To cte ths verson: Jean-Marc Vannobel. Contnuous Belef Functons: Focal Intervals Propertes.. BELIEF 212, May 212, Compègne, France.
More informationAPPENDIX A Some Linear Algebra
APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,
More informationSalmon: Lectures on partial differential equations. Consider the general linear, second-order PDE in the form. ,x 2
Salmon: Lectures on partal dfferental equatons 5. Classfcaton of second-order equatons There are general methods for classfyng hgher-order partal dfferental equatons. One s very general (applyng even to
More information: Numerical Analysis Topic 2: Solution of Nonlinear Equations Lectures 5-11:
764: Numercal Analyss Topc : Soluton o Nonlnear Equatons Lectures 5-: UIN Malang Read Chapters 5 and 6 o the tetbook 764_Topc Lecture 5 Soluton o Nonlnear Equatons Root Fndng Problems Dentons Classcaton
More informationCS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang
CS DESIGN ND NLYSIS OF LGORITHMS DYNMIC PROGRMMING Dr. Dasy Tang Dynamc Programmng Idea: Problems can be dvded nto stages Soluton s a sequence o decsons and the decson at the current stage s based on the
More informationFault Detection and Diagnosis in Dynamic Systems with Maximal Sensitivity. S. Joe Qin and Weihua Li*
Fault Detecton and Dagnoss n Dynamc Systems wth Maxmal Senstvty S. Joe Qn and Wehua L* Department o Chemcal Engneerng he Unversty o exas at Austn Austn, exas 787 5-7-7 qn@che.utexas.edu Control.che.utexas.edu/qnlab
More informationAutomatica. Invertibility of switched nonlinear systems. Aneel Tanwani, Daniel Liberzon. a b s t r a c t. 1. Introduction
Automatca 46 (2010) 1962 1973 Contents lsts avalable at ScenceDrect Automatca journal homepage: wwwelsevercom/locate/automatca Invertblty o swtched nonlnear systems Aneel Tanwan, Danel Lberzon Coordnated
More informationQuantum Mechanics I - Session 4
Quantum Mechancs I - Sesson 4 Aprl 3, 05 Contents Operators Change of Bass 4 3 Egenvectors and Egenvalues 5 3. Denton....................................... 5 3. Rotaton n D....................................
More information= s j Ui U j. i, j, then s F(U) with s Ui F(U) G(U) F(V ) G(V )
1 Lecture 2 Recap Last tme we talked about presheaves and sheaves. Preshea: F on a topologcal space X, wth groups (resp. rngs, sets, etc.) F(U) or each open set U X, wth restrcton homs ρ UV : F(U) F(V
More informationLinear Approximation with Regularization and Moving Least Squares
Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...
More informationNumerical Solutions of a Generalized Nth Order Boundary Value Problems Using Power Series Approximation Method
Appled Mathematcs, 6, 7, 5-4 Publshed Onlne Jul 6 n ScRes. http://www.scrp.org/journal/am http://.do.org/.436/am.6.77 umercal Solutons of a Generalzed th Order Boundar Value Problems Usng Power Seres Approxmaton
More informationLectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix
Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could
More informationInexact Newton Methods for Inverse Eigenvalue Problems
Inexact Newton Methods for Inverse Egenvalue Problems Zheng-jan Ba Abstract In ths paper, we survey some of the latest development n usng nexact Newton-lke methods for solvng nverse egenvalue problems.
More informationDesigning of Combined Continuous Lot By Lot Acceptance Sampling Plan
Internatonal Journal o Scentc Research Engneerng & Technology (IJSRET), ISSN 78 02 709 Desgnng o Combned Contnuous Lot By Lot Acceptance Samplng Plan S. Subhalakshm 1 Dr. S. Muthulakshm 2 1 Research Scholar,
More informationNeuro-Adaptive Design - I:
Lecture 36 Neuro-Adaptve Desgn - I: A Robustfyng ool for Dynamc Inverson Desgn Dr. Radhakant Padh Asst. Professor Dept. of Aerospace Engneerng Indan Insttute of Scence - Bangalore Motvaton Perfect system
More informationErrors for Linear Systems
Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch
More informationEE 330 Lecture 24. Small Signal Analysis Small Signal Analysis of BJT Amplifier
EE 0 Lecture 4 Small Sgnal Analss Small Sgnal Analss o BJT Ampler Eam Frda March 9 Eam Frda Aprl Revew Sesson or Eam : 6:00 p.m. on Thursda March 8 n Room Sweene 6 Revew rom Last Lecture Comparson o Gans
More informationMATH 829: Introduction to Data Mining and Analysis The EM algorithm (part 2)
1/16 MATH 829: Introducton to Data Mnng and Analyss The EM algorthm (part 2) Domnque Gullot Departments of Mathematcal Scences Unversty of Delaware Aprl 20, 2016 Recall 2/16 We are gven ndependent observatons
More informationTime-Varying Systems and Computations Lecture 6
Tme-Varyng Systems and Computatons Lecture 6 Klaus Depold 14. Januar 2014 The Kalman Flter The Kalman estmaton flter attempts to estmate the actual state of an unknown dscrete dynamcal system, gven nosy
More informationON FIBRANT OBJECTS IN MODEL CATEGORIES
Theory and Applcatons o Categores, ol. 33, No. 3, 2018, pp. 43 66. ON FIBRANT OBJECTS IN MODEL CATEGORIES ALERY ISAE Abstract. In ths paper, we study propertes o maps between brant objects n model categores.
More informationRobust Sliding Mode Observers for Large Scale Systems with Applications to a Multimachine Power System
Robust Sldng Mode Observers for Large Scale Systems wth Applcatons to a Multmachne Power System Mokhtar Mohamed, Xng-Gang Yan,*, Sarah K. Spurgeon 2, Bn Jang 3 Instrumentaton, Control and Embedded Systems
More informationn-strongly Ding Projective, Injective and Flat Modules
Internatonal Mathematcal Forum, Vol. 7, 2012, no. 42, 2093-2098 n-strongly Dng Projectve, Injectve and Flat Modules Janmn Xng College o Mathematc and Physcs Qngdao Unversty o Scence and Technology Qngdao
More informationNumerical Heat and Mass Transfer
Master degree n Mechancal Engneerng Numercal Heat and Mass Transfer 06-Fnte-Dfference Method (One-dmensonal, steady state heat conducton) Fausto Arpno f.arpno@uncas.t Introducton Why we use models and
More informationFirst day August 1, Problems and Solutions
FOURTH INTERNATIONAL COMPETITION FOR UNIVERSITY STUDENTS IN MATHEMATICS July 30 August 4, 997, Plovdv, BULGARIA Frst day August, 997 Problems and Solutons Problem. Let {ε n } n= be a sequence of postve
More informationOn a direct solver for linear least squares problems
ISSN 2066-6594 Ann. Acad. Rom. Sc. Ser. Math. Appl. Vol. 8, No. 2/2016 On a drect solver for lnear least squares problems Constantn Popa Abstract The Null Space (NS) algorthm s a drect solver for lnear
More informationLecture 21: Numerical methods for pricing American type derivatives
Lecture 21: Numercal methods for prcng Amercan type dervatves Xaoguang Wang STAT 598W Aprl 10th, 2014 (STAT 598W) Lecture 21 1 / 26 Outlne 1 Fnte Dfference Method Explct Method Penalty Method (STAT 598W)
More informationCISE301: Numerical Methods Topic 2: Solution of Nonlinear Equations
CISE3: Numercal Methods Topc : Soluton o Nonlnear Equatons Dr. Amar Khoukh Term Read Chapters 5 and 6 o the tetbook CISE3_Topc c Khoukh_ Lecture 5 Soluton o Nonlnear Equatons Root ndng Problems Dentons
More informationCSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography
CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve
More informationParametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010
Parametrc fractonal mputaton for mssng data analyss Jae Kwang Km Survey Workng Group Semnar March 29, 2010 1 Outlne Introducton Proposed method Fractonal mputaton Approxmaton Varance estmaton Multple mputaton
More informationECEN 5005 Crystals, Nanocrystals and Device Applications Class 19 Group Theory For Crystals
ECEN 5005 Crystals, Nanocrystals and Devce Applcatons Class 9 Group Theory For Crystals Dee Dagram Radatve Transton Probablty Wgner-Ecart Theorem Selecton Rule Dee Dagram Expermentally determned energy
More informationYong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 )
Kangweon-Kyungk Math. Jour. 4 1996), No. 1, pp. 7 16 AN ITERATIVE ROW-ACTION METHOD FOR MULTICOMMODITY TRANSPORTATION PROBLEMS Yong Joon Ryang Abstract. The optmzaton problems wth quadratc constrants often
More informationGroup Analysis of Ordinary Differential Equations of the Order n>2
Symmetry n Nonlnear Mathematcal Physcs 997, V., 64 7. Group Analyss of Ordnary Dfferental Equatons of the Order n> L.M. BERKOVICH and S.Y. POPOV Samara State Unversty, 4430, Samara, Russa E-mal: berk@nfo.ssu.samara.ru
More informationSome Star and Bistar Related Divisor Cordial Graphs
Annals o Pure and Appled Mathematcs Vol. 3 No. 03 67-77 ISSN: 79-087X (P) 79-0888(onlne) Publshed on 3 May 03 www.researchmathsc.org Annals o Some Star and Bstar Related Dvsor Cordal Graphs S. K. Vadya
More informationSTAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 16
STAT 39: MATHEMATICAL COMPUTATIONS I FALL 218 LECTURE 16 1 why teratve methods f we have a lnear system Ax = b where A s very, very large but s ether sparse or structured (eg, banded, Toepltz, banded plus
More informationJournal of Universal Computer Science, vol. 1, no. 7 (1995), submitted: 15/12/94, accepted: 26/6/95, appeared: 28/7/95 Springer Pub. Co.
Journal of Unversal Computer Scence, vol. 1, no. 7 (1995), 469-483 submtted: 15/12/94, accepted: 26/6/95, appeared: 28/7/95 Sprnger Pub. Co. Round-o error propagaton n the soluton of the heat equaton by
More informationComputational Biology Lecture 8: Substitution matrices Saad Mneimneh
Computatonal Bology Lecture 8: Substtuton matrces Saad Mnemneh As we have ntroduced last tme, smple scorng schemes lke + or a match, - or a msmatch and -2 or a gap are not justable bologcally, especally
More informationImplicit Integration Henyey Method
Implct Integraton Henyey Method In realstc stellar evoluton codes nstead of a drect ntegraton usng for example the Runge-Kutta method one employs an teratve mplct technque. Ths s because the structure
More informationCHAPTER 4d. ROOTS OF EQUATIONS
CHAPTER 4d. ROOTS OF EQUATIONS A. J. Clark School o Engneerng Department o Cvl and Envronmental Engneerng by Dr. Ibrahm A. Assakka Sprng 00 ENCE 03 - Computaton Methods n Cvl Engneerng II Department o
More informationDeriving the X-Z Identity from Auxiliary Space Method
Dervng the X-Z Identty from Auxlary Space Method Long Chen Department of Mathematcs, Unversty of Calforna at Irvne, Irvne, CA 92697 chenlong@math.uc.edu 1 Iteratve Methods In ths paper we dscuss teratve
More informationThe Geometry of Logit and Probit
The Geometry of Logt and Probt Ths short note s meant as a supplement to Chapters and 3 of Spatal Models of Parlamentary Votng and the notaton and reference to fgures n the text below s to those two chapters.
More informationThe KMO Method for Solving Non-homogenous, m th Order Differential Equations
The KMO Method for Solvng Non-homogenous, m th Order Dfferental Equatons Davd Krohn Danel Marño-Johnson John Paul Ouyang March 14, 2013 Abstract Ths paper shows a smple tabular procedure for fndng the
More informationCanonical transformations
Canoncal transformatons November 23, 2014 Recall that we have defned a symplectc transformaton to be any lnear transformaton M A B leavng the symplectc form nvarant, Ω AB M A CM B DΩ CD Coordnate transformatons,
More information2.3 Nilpotent endomorphisms
s a block dagonal matrx, wth A Mat dm U (C) In fact, we can assume that B = B 1 B k, wth B an ordered bass of U, and that A = [f U ] B, where f U : U U s the restrcton of f to U 40 23 Nlpotent endomorphsms
More informationGlobal Sensitivity. Tuesday 20 th February, 2018
Global Senstvty Tuesday 2 th February, 28 ) Local Senstvty Most senstvty analyses [] are based on local estmates of senstvty, typcally by expandng the response n a Taylor seres about some specfc values
More informationProblem Set 9 Solutions
Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem
More informationform, and they present results of tests comparng the new algorthms wth other methods. Recently, Olschowka & Neumaer [7] ntroduced another dea for choo
Scalng and structural condton numbers Arnold Neumaer Insttut fur Mathematk, Unverstat Wen Strudlhofgasse 4, A-1090 Wen, Austra emal: neum@cma.unve.ac.at revsed, August 1996 Abstract. We ntroduce structural
More informationComplete subgraphs in multipartite graphs
Complete subgraphs n multpartte graphs FLORIAN PFENDER Unverstät Rostock, Insttut für Mathematk D-18057 Rostock, Germany Floran.Pfender@un-rostock.de Abstract Turán s Theorem states that every graph G
More informationAffine transformations and convexity
Affne transformatons and convexty The purpose of ths document s to prove some basc propertes of affne transformatons nvolvng convex sets. Here are a few onlne references for background nformaton: http://math.ucr.edu/
More informationBitopological spaces via Double topological spaces
topologcal spaces va Double topologcal spaces KNDL O TNTWY SEl-Shekh M WFE Mathematcs Department Faculty o scence Helwan Unversty POox 795 aro Egypt Mathematcs Department Faculty o scence Zagazg Unversty
More informationLinear Algebra and its Applications
Lnear Algebra and ts Applcatons 4 (00) 5 56 Contents lsts avalable at ScenceDrect Lnear Algebra and ts Applcatons journal homepage: wwwelsevercom/locate/laa Notes on Hlbert and Cauchy matrces Mroslav Fedler
More informationCME 302: NUMERICAL LINEAR ALGEBRA FALL 2005/06 LECTURE 13
CME 30: NUMERICAL LINEAR ALGEBRA FALL 005/06 LECTURE 13 GENE H GOLUB 1 Iteratve Methods Very large problems (naturally sparse, from applcatons): teratve methods Structured matrces (even sometmes dense,
More informationSL n (F ) Equals its Own Derived Group
Internatonal Journal of Algebra, Vol. 2, 2008, no. 12, 585-594 SL n (F ) Equals ts Own Derved Group Jorge Macel BMCC-The Cty Unversty of New York, CUNY 199 Chambers street, New York, NY 10007, USA macel@cms.nyu.edu
More informationFinite time observation of nonlinear time-delay systems with unknown inputs
Author manuscript, published in "NOLCOS21 (21)" Finite time observation o nonlinear time-delay systems with unknown inputs G. Zheng J.-P. Barbot, D. Boutat, T. Floquet, J.-P. Richard INRIA Lille-Nord Europe,
More informationAn efficient algorithm for multivariate Maclaurin Newton transformation
Annales UMCS Informatca AI VIII, 2 2008) 5 14 DOI: 10.2478/v10065-008-0020-6 An effcent algorthm for multvarate Maclaurn Newton transformaton Joanna Kapusta Insttute of Mathematcs and Computer Scence,
More informationExample: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41,
The greatest common dvsor of two ntegers a and b (not both zero) s the largest nteger whch s a common factor of both a and b. We denote ths number by gcd(a, b), or smply (a, b) when there s no confuson
More informationMaximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models
ECO 452 -- OE 4: Probt and Logt Models ECO 452 -- OE 4 Maxmum Lkelhood Estmaton of Bnary Dependent Varables Models: Probt and Logt hs note demonstrates how to formulate bnary dependent varables models
More informationn α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0
MODULE 2 Topcs: Lnear ndependence, bass and dmenson We have seen that f n a set of vectors one vector s a lnear combnaton of the remanng vectors n the set then the span of the set s unchanged f that vector
More informationLINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity
LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 31 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 6. Rdge regresson The OLSE s the best lnear unbased
More informationChat eld, C. and A.J.Collins, Introduction to multivariate analysis. Chapman & Hall, 1980
MT07: Multvarate Statstcal Methods Mke Tso: emal mke.tso@manchester.ac.uk Webpage for notes: http://www.maths.manchester.ac.uk/~mkt/new_teachng.htm. Introducton to multvarate data. Books Chat eld, C. and
More informationA Robust Method for Calculating the Correlation Coefficient
A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal
More informationMEM 255 Introduction to Control Systems Review: Basics of Linear Algebra
MEM 255 Introducton to Control Systems Revew: Bascs of Lnear Algebra Harry G. Kwatny Department of Mechancal Engneerng & Mechancs Drexel Unversty Outlne Vectors Matrces MATLAB Advanced Topcs Vectors A
More informationLecture 10 Support Vector Machines II
Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed
More informationBOUNDEDNESS OF THE RIESZ TRANSFORM WITH MATRIX A 2 WEIGHTS
BOUNDEDNESS OF THE IESZ TANSFOM WITH MATIX A WEIGHTS Introducton Let L = L ( n, be the functon space wth norm (ˆ f L = f(x C dx d < For a d d matrx valued functon W : wth W (x postve sem-defnte for all
More informationMathematical Preparations
1 Introducton Mathematcal Preparatons The theory of relatvty was developed to explan experments whch studed the propagaton of electromagnetc radaton n movng coordnate systems. Wthn expermental error the
More informationComposite Hypotheses testing
Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter
More informationComputing Correlated Equilibria in Multi-Player Games
Computng Correlated Equlbra n Mult-Player Games Chrstos H. Papadmtrou Presented by Zhanxang Huang December 7th, 2005 1 The Author Dr. Chrstos H. Papadmtrou CS professor at UC Berkley (taught at Harvard,
More informationSUCCESSIVE MINIMA AND LATTICE POINTS (AFTER HENK, GILLET AND SOULÉ) M(B) := # ( B Z N)
SUCCESSIVE MINIMA AND LATTICE POINTS (AFTER HENK, GILLET AND SOULÉ) S.BOUCKSOM Abstract. The goal of ths note s to present a remarably smple proof, due to Hen, of a result prevously obtaned by Gllet-Soulé,
More informationLecture 4: Universal Hash Functions/Streaming Cont d
CSE 5: Desgn and Analyss of Algorthms I Sprng 06 Lecture 4: Unversal Hash Functons/Streamng Cont d Lecturer: Shayan Oves Gharan Aprl 6th Scrbe: Jacob Schreber Dsclamer: These notes have not been subjected
More informationFACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP
C O L L O Q U I U M M A T H E M A T I C U M VOL. 80 1999 NO. 1 FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP BY FLORIAN K A I N R A T H (GRAZ) Abstract. Let H be a Krull monod wth nfnte class
More informationSnce h( q^; q) = hq ~ and h( p^ ; p) = hp, one can wrte ~ h hq hp = hq ~hp ~ (7) the uncertanty relaton for an arbtrary state. The states that mnmze t
8.5: Many-body phenomena n condensed matter and atomc physcs Last moded: September, 003 Lecture. Squeezed States In ths lecture we shall contnue the dscusson of coherent states, focusng on ther propertes
More informationCOMPLEX NUMBERS AND QUADRATIC EQUATIONS
COMPLEX NUMBERS AND QUADRATIC EQUATIONS INTRODUCTION We know that x 0 for all x R e the square of a real number (whether postve, negatve or ero) s non-negatve Hence the equatons x, x, x + 7 0 etc are not
More informationOn simultaneous parameter identification and state estimation for cascade state affine systems
Amercan Control Conference Westn Seattle Hotel, Seattle, Washngton, USA June 11-13, WeAI1.9 On smultaneous parameter dentfcaton and state estmaton for cascade state affne systems M. GHANES, G. ZHENG and
More informationObservers for non-linear differential-algebraic. system
Observers for non-lnear dfferental-algebrac systems Jan Åslund and Erk Frsk Department of Electrcal Engneerng, Lnköpng Unversty 581 83 Lnköpng, Sweden, {jaasl,frsk}@sy.lu.se Abstract In ths paper we consder
More informationAbsorbing Markov Chain Models to Determine Optimum Process Target Levels in Production Systems with Rework and Scrapping
Archve o SID Journal o Industral Engneerng 6(00) -6 Absorbng Markov Chan Models to Determne Optmum Process Target evels n Producton Systems wth Rework and Scrappng Mohammad Saber Fallah Nezhad a, Seyed
More information5 The Rational Canonical Form
5 The Ratonal Canoncal Form Here p s a monc rreducble factor of the mnmum polynomal m T and s not necessarly of degree one Let F p denote the feld constructed earler n the course, consstng of all matrces
More informationThe Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction
ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also
More informationLocal Approximation of Pareto Surface
Proceedngs o the World Congress on Engneerng 007 Vol II Local Approxmaton o Pareto Surace S.V. Utyuzhnkov, J. Magnot, and M.D. Guenov Abstract In the desgn process o complex systems, the desgner s solvng
More information9 Characteristic classes
THEODORE VORONOV DIFFERENTIAL GEOMETRY. Sprng 2009 [under constructon] 9 Characterstc classes 9.1 The frst Chern class of a lne bundle Consder a complex vector bundle E B of rank p. We shall construct
More informationCooperative Output Regulation of Linear Multi-agent Systems with Communication Constraints
2016 IEEE 55th Conference on Decson and Control (CDC) ARIA Resort & Casno December 12-14, 2016, Las Vegas, USA Cooperatve Output Regulaton of Lnear Mult-agent Systems wth Communcaton Constrants Abdelkader
More informationConvexity preserving interpolation by splines of arbitrary degree
Computer Scence Journal of Moldova, vol.18, no.1(52), 2010 Convexty preservng nterpolaton by splnes of arbtrary degree Igor Verlan Abstract In the present paper an algorthm of C 2 nterpolaton of dscrete
More informationUniversity of Washington Department of Chemistry Chemistry 452/456 Summer Quarter 2014
Lecture 16 8/4/14 Unversty o Washngton Department o Chemstry Chemstry 452/456 Summer Quarter 214. Real Vapors and Fugacty Henry s Law accounts or the propertes o extremely dlute soluton. s shown n Fgure
More informationUsing T.O.M to Estimate Parameter of distributions that have not Single Exponential Family
IOSR Journal of Mathematcs IOSR-JM) ISSN: 2278-5728. Volume 3, Issue 3 Sep-Oct. 202), PP 44-48 www.osrjournals.org Usng T.O.M to Estmate Parameter of dstrbutons that have not Sngle Exponental Famly Jubran
More informationPHYS 705: Classical Mechanics. Canonical Transformation II
1 PHYS 705: Classcal Mechancs Canoncal Transformaton II Example: Harmonc Oscllator f ( x) x m 0 x U( x) x mx x LT U m Defne or L p p mx x x m mx x H px L px p m p x m m H p 1 x m p m 1 m H x p m x m m
More informationSolutions to exam in SF1811 Optimization, Jan 14, 2015
Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable
More information1 GSW Iterative Techniques for y = Ax
1 for y = A I m gong to cheat here. here are a lot of teratve technques that can be used to solve the general case of a set of smultaneous equatons (wrtten n the matr form as y = A), but ths chapter sn
More information