Identication
WolfgangReinelt #
,AndreaGarulli
∗
andLennartLjung #
#
Department ofElectrical Engineering
Linkoping University, 58183Linkoping, Sweden
WWW:
http://www.control.isy.liu.se/
E-mail:
{wolle,ljung}@isy.liu.se
∗
Dipartimentodi Ingegneriadell'Informazione
Universita di Siena,53100Siena,Italy
WWW:
http://www-dii.ing.unisi.it/~garulli/
E-mail:
garulli@ing.unisi.it
August2, 2000
REG
LERTEKNIK
AUTO
MATIC CONTR
OL
LINKÖPING
Reportno.: LiTH-ISY-R-2293
PreparedforJournal Publication
Technical reports from the Automatic Control group in Linkoping are available by
anony-mous ftp at the address
ftp.control.isy.liu.se
. This report is contained inthe portablein Robust Identication
∗
Wolfgang Reinelt #,
†
,Andrea Garulli ##andLennartLjung #
#
DepartmentofElectricalEngineering,
Linkoping University,58183 Linkoping,Sweden.
E-mail:
{wolle,ljung}@isy.liu.se
##
Dipartimentodi Ingegneriadell'Informazione,
Universita di Siena,53100Siena, Italy.
E-mail:
garulli@ing.unisi.it
Abstract
Identicationforrobustcontrolmustdelivernotonlyanominalmodel,butalsoa
re-liableestimateoftheuncertaintyassociatedwiththemodel. Thispaperaddressesrecent
approachestorobustidentication, thatexplicitly aimatseparatingcontributionsfrom
thetwomainuncertaintysources: unmodeled dynamicsandnoiseaectingthedata. In
particular,thefollowing methodsareconsidered: non-stationaryStochasticEmbedding,
Model ErrorModeling based onprediction error methods,Set Membership
Identica-tion. Modelvalidationissuesarealsoaddressedintheproposedframework. Moreover,
itisshownhowthecomputationoftheminimumnoiseboundforwhichanominalmodel
isnotfalsied byi/odata,canbeusedasarationale forselectinganappropriatemodel
class in thesetmembershipsetting. For allmethods,uncertainty isevaluatedin terms
ofthefrequencyresponse,sothatitcanbehandledby
H∞
controltechniques. An exam-ple,whereanontrivialundermodelingisensuredbythepresenceofanonlinearityinthesystemgeneratingthedata,ispresentedtocomparethedierentmethods.
Keywords: identication for robust control, model error modeling, unmodeled dynamics,
modelvalidation,setmembershipestimation,stochasticembedding.
∗
Preliminary versions of thiswork have been presented atthe1999 Conference on Decisionand Control,
Phoenix,AZ,USA[19]andtheIFACSymposiumSystemIdenticationSYSID2000,SantaBarbara,CA,USA[6 ].
†
SupportedbytheEuropeanUnionwithintheEuropeanResearchNetworkinSystemIdentication(ERNSI)
Oneof themainobjectivesof control-orientedidenticationistoestimatemodelsthatare
suitable for robust control design techniques. For this purpose, the identication
proce-duremustdelivernotonly anominalmodel,butalsoareliableestimateoftheuncertainty
associated tothe model. Dierent paradigms forthe description ofuncertainty have been
addressedintheliterature(seee.g.thespecialissues[13,20]andtherecentbook[7]). Two
main philosophies are basically adopted. The rst one is based on statistical assumptions
andusuallyleadstoleastsquares estimationtechniquesand predictionerror methods,see
[16]. The second one relies on deterministic hypotheses, such as the identication error
being unknown-but-bounded (UBB), and has givenrise to a numberof techniques usually
addressedasboundederrororsetmembershipidentication(see[18,17]andthereferences
therein).
In standard identication problems, the error originates from two dierent sources: a
\variance"term,duetonoiseaectingthedata,anda\bias"term,duetosystemdynamics
whichis notcaptured by the estimatednominalmodel (oftenaddressed alsoas the model
error). Clearly,thenatureofthesetwoerrortermsisquitedierent: theformerisgenerally
uncorrelatedwiththeinputsignal(whenthedatais collectedinopenloop),whilethelatter
stronglydependsontheestimatednominalmodelandontheinputusedintheidentication
experiment. The model erroris notnegligible inmost practical situations,especiallythose
in which the order of the nominal model must be small (a typical requirement of robust
design techniques). Moreover, whileapriori informationon measurementnoise areoften
available,similarhypothesesontheunmodeleddynamicsseemtobelessrealistic.
This paperaddressesthreedierent approachestorobustidentication: Stochastic
Em-bedding,Model ErrorModelingandSetMembershipIdentication. Eachofthemexplicitly
separatecontributionsfrom theabovetwoerror sources,thusprovidingareliableestimate
ofmodeluncertainty. Thersttwoapproacheshavebeendevelopedinthestatistical
frame-work,whilethelatterrelieson UBBerror assumptions.
StochasticEmbedding[12,10]isafrequencydomainmethodwhichassumesthat
unmod-eled dynamicscanberepresentedadequatelyby anon-stationary stochasticprocesswhose
varianceincreaseswithfrequency. Thenominalmodelisobtainedvialeast-squares
estima-tionfromfrequencydomaindata; therefore, harmonicinputsarerequired. Theuncertainty
associatedtothemodelis evaluatedfrom statisticalpropertiesoftherandomwalkprocess
describingtheunmodeleddynamics.
Model Error Modeling [14, 15] employs standard prediction error methods to identify
a nominal model from input-output time domain data [16]. Then, one can estimate the
unmodeleddynamicsbylookingatthatpartoftheidenticationresidualthatoriginatesfrom
theinput. Identicationofresidualdynamics(thatcanbeperformedusingagainprediction
error methods) provides theso called model errormodel. The condenceregionof model
SetMembershipIdenticationprovideseÆcientalgorithmsforestimatingthesetof
fea-siblemodels,compatiblewiththeavailabledataandtheUBBerror assumption. Thechoice
ofthenominalmodelis usuallyperformedbyminimizingacostfunctionrelatedtothe
fea-sible set. The feasiblesetitselfgives the sizeof the uncertainty associatedto the nominal
model. Intherstworksonsetmembershipidentication(seethesurvey[25]andthe
ref-erencestherein)thecontributions fromunmodeleddynamics andnoisewerenotseparated
(which corresponds to assuming that the plant generating the data belongs to the
consid-eredmodelclass).Morerecently,thepresenceofmodelerrorshasbeenexplicitlyaccounted
for inseveral works and dierent settings (seee.g. [24, 9, 8]). In thispaper, modelerror
modelingconceptsareextendedtosetmembershipidentication,inordertoobtainafairly
generalstrategyforseparatingerrorcontributions.
Asitcanbeseenfromtheabovediscussion,quitedierentapproachestorobust
identi-cationareprovidedbytheconsideredtechniques,themaindierencesconcerningthetype
of data required, theselectedmodelclass, the assumptionson thenoise aectingthe data
andthenominalmodelestimationcriteria. Themaincontributionofthepaperistotestthe
abovethreeapproachesonthesamesimulationsetting,inordertohighlighttheadvantages
ofeachmethodandtoprovide,aslongasitispossible,afaircomparisonoftheresults. The
mostimportantfeaturesofeachtechniqueareemphasized. In ordertoensurethepresence
ofanontrivialundermodeling,anonlinearityispresentinthetruesystemconsideredinthe
test example, and identicationof a low order nominal model is required. Frequency
do-mainuncertaintyregionsareconsideredinordertoprovideadequatemodelsfor
H
∞
control design.The secondcontributionofthepaperistoprovideafrequencydomainmodelvalidation
tool forsetmembershipidentication. The basicidea is toevaluatethe minimumvalue of
theerrorbound,whichallowsonetovalidatetheestimatednominalmodel. Itisalsoshown
thatthisprovidesausefulrationalefortheselectionofanadequatemodelclass.
Paper outline: Section 2 brie yreviews the technique ofnon-stationary stochastic
embed-ding, while in Section3, modelerror modeling ideas aresummarized from aquite general
point of view. Section 4 introduces the concept of set membership identication,
embed-ded in a model error modeling setup, and discusses identication of both nominal model
and model error, modelvalidation and model classselection. Section 5reports a
compar-ative simulation example, which illustrates the main features of non-stationary stochastic
embedding and model error modeling, the latter using eitherprediction error methods or
theproposedsetmembershipidenticationapproach. Benets andadvantagesofthethree
techniques are discussed. Finally, Section 6 gives some concluding remarks and suggests
ThebasicideabehindStochasticEmbeddingcanbedescribedasfollows. Thetruesystem
G
canbegivenas
G(iω) = ^
G
0
(iω) + ∆G(iω)
(1)where
G
^
0
is a \nominal system" that can be exactly represented within a parameterizedfamily,and
∆G(iω)
isarandomvariableindependentof^
G
0
. Noticethatthisisaturn-around oftheconventionalinterpretation,wherethenominal,estimatedmodel^
G
0
(iω) = G(iω) + ∆G(iω)
is seen as the true system plus a model error
∆G
, that is independent ofG
. In contrast,thetruesystemin(1)isarandomvariableas well,whichistherootoftheterm\stochastic
embedding". Now,supposethatwehavenoisyobservationsofthetruesystem
G
atcertainfrequencies:
^
^
G
k
= G(iω
k
) + ν
k
(2)wherethenoiseterm
ν
k
isindependentofG
and∆G
. Then,combining(1)and(2),onehas^
^
G
k
= ^
G
0
(iω
k
) + ν
k
+ ∆G(iω
k
).
(3)Thisis thebasicestimationequation,thatneedstobecomplementedwitha
parameteriza-tion ofthe model
G
^
0
and assumptions about the variance ofν
k
and∆G(iω
k
)
. In [10]it issuggestedtousealinearregressionparameterizationofthenominalmodelintermsofsome
orthonormal basis functions,i.e.
G
^
0
(θ) =
P
n
i=1
θ
i
B
i
, whereB := [B
1
, . . . , B
n
]
represents the selectedbasis andθ
is thevectorofparameters. Moreover, themodelerror∆G
isparame-terized according tothe samebasis
B
and itis assumedthat the relativemodel error hasa variancethatincreaseslinearlyorquadraticallywithω
. Thismeansthat∆G
canbewrittenas
B
θΛ
whereΛ
is a random walk process overω
(this argument being suppressed) andθ
comesfrompriorknowledge. AllthisleadstoG
=
G
^
0
(θ) + G
0
(
θ)Λ
(4)
=
Bθ + B
θΛ.
(5)
Theidenticationprocedurethenconsistsofthreemainsteps:
1. Pointwiseleastsquaresestimationofthetransferfunctionforcertainfrequencies:
there-fore,theinput
u
hastobeasumofsinusoids. Thisstepdeliversthevalues^
^
G
k
,forcer-tainfrequencies. Additionally, statisticalproperties ofthe noise
v
are calculated(i.e.anunbiasedestimateofitsvariance),assumingGaussiannoise.
3. Estimation ofthe parameter
θ
and therandomwalk processΛ
in eqn.(5):^
^
G
k
=
Bθ +
B
θΛ+ν
k
,basedonthefrequencyfunctionpointestimates^
^
G
k
. Thisisusuallyperformed accordingtothefollowingprocedure:(a) Estimationof
θ
(calledθ
^
),basedontheknowledgeof^
^
G
k
. Thus,thenominalmodel^
G
0
=
B^θ
ineqn.(4) is the leastsquares estimateapproximationof^
^
G
k
inthesub-spacespannedbybasisfunctions
B
.(b) Asmodelerrorparameterization
θ
,theestimateθ
^
ischosen(thisisatypicalchoice, whennoaprioriinformationonunmodeleddynamicsis available). Thus,eqn.(5)becomes:
^
^
G
k
=
B^θ(1 + Λ) + ν
k
.
(c) Computationofanunbiasedestimateofthevarianceoftherandomwalk
Λ
,basi-callyusing anunbiased estimate ofthe varianceofthe noise, see[10] forclosed
formexpressions and technicaldetails. Here,the randomwalkis amodel whose
varianceofthefrequencyresponseincreaseslinearlywithfrequencyuptoacertain
rollo.
(d) Quantication of the model error
B
θΛ
for any frequency, i.e. calculation of its statisticalproperties.Avariantoftherandomwalkforresonant systems(integratedrandomwalk)isdescribedin
[2]: here, the error increases quadraticallywith the frequency, in order to produce tighter
error boundsat lowfrequenciesinthepresenceofresonant poles.
Step2containsthecrucialchoiceofthesetof(orthonormal)basisfunctions. Inthesimple
caseofaLaguerreexpansion,onecanemploytheminimizationoftheaveragemodelingerror,
as suggestedin[1, 2]. Atanyfrequency
ω
j
, the total modeling error is givenbyG
e
(ω
j
) :=
B(ω
j
)^
θ−G(ω
j
)
. ForaclosedformexpressionofG
e
(ω
j
)
anditscovarianceΣ
e
(ω
j
)
,thereaderisreferredto[11 ]. Wenotethat,fordierentchoicesofthepoleintheLaguerreexpansion,the
covariancematrixisafunctionoftheLaguerrepoleaswell:
Σ
e
(ω
j
) = Σ
e
(ω
j
, p
i
)
. Therfore,wedenetheaveragemodelingerror(overacertainnitefrequencygrid
Ω
)foraxedLaguerrepoleas
X
ω
j
∈Ω
trace
(Σ
e
(ω
j
, p
i
))
(6)The\optimal"Laguerrepole
p
∗
i
istheonethatminimizesthemeanerror(6).3 Model Error Modeling
Inthissection,thebasicideasofmodelerrormodelingaredescribed. Athoroughtreatment
of this approach is given in [15] , where prediction error methods are used to calculate
nominal model as well as error model. The idea described there, however, is muchmore
Let
(u
m
, y
m
)
beacollectionof i/o measurements and assumethata nominalmodelG
n
ofthe systemthat generatedthedatahas beenestimated, accordingtosome identication
procedure. Then,themodelerror modelingstrategycanbesummarizedasfollows.
1. Computetheresidual
= y
m
− G
n
u
m
.2. Consider the \error" system, with input
u
m
and output , and identify a modelG
e
for thissystem. This is an estimateof the error due toundermodeling, the so-called
modelerrormodel. Standardstatisticalpropertiesoftheestimatedmodelleadtothe
denitionoftheuncertaintyregionofthemodelerror(e.g.,givenbythe99%condence
regionoftheestimatedmodelerror
G
e
).3. The uncertainty region of the nominal model is obtained by adding up the nominal
modelandtheuncertaintyregionofthemodelerror.
4. Model validation: the nominal model is not falsied if and only if it lies inside its
ownuncertaintyregionor,equivalently,ifand onlyif zerois anelementofthemodel
error uncertainty region. This can beeasily checked by lookingat the corresponding
uncertaintybandsintheBodeorNyquistplots.
Identicationofthemodelerrorfromresidualdataprovidesaseparationbetweennoise
andunmodeleddynamics. Infact,
G
e
canbeseenasanestimateofthedynamicsystem∆(
·)
(possiblynonlinear,time-varying,etc.), suchthat= ∆(u) + v
(7)where
v
isthenoiseterm,whichis assumedtobeuncorrelatedwiththeinputu
.Afewobservationscanbemade.
•
The modelerrorG
e
must beselectedamong apre-speciedclassofmodels. Unfortu-nately,thereisnotastandardprocedureforselectingthestructureofthemodelerror.Thisisratherbasedonaprioriknowledgeonthesystemgeneratingthedata,andonthe
purposesforwhichthemodeluncertaintyisestimated. Forexample,atypical
require-mentis thatunmodeled dynamicsis capturedat thosefrequencies thataresignicant
forcontroldesignpurposes.
•
Thesizeoftheuncertaintyregionclearlydependsontheselectednominalmodel. When a severe undermodeling occurs (due for example to the fact that we are required todelivera loworder nominalmodeltothe robust controldesigner), one cannotexpect
themodelerroruncertaintyregiontobesmallatallfrequencies,i.e. thenominalmodel
willbefalsied,whateverthestructureofthemodelerroris.
•
Afalsiednominalmodelmaystillbeaccepted, provided thattherelateduncertainty bandissmallatfrequenciesofinterest.In thispaper, wewill adopt the general classof linearmodels proposed in[16] and apply
4.1 BasicIdeasand connectiontoModel ErrorModeling
ThetermSetMembershipIdenticationhasbeenusedinrecentyearstoindicateawide
va-rietyofrobustidenticationtechniques,thatareabletohandlehardboundsonthe
identi-cationerror. Theapproachconsideredinthispaperistheonetakenin[8]. Inthefollowing,
wesummarizeits mainfeatures.
Assumethat themeasuredi/odataaregeneratedby asystem
S
0
,accordingtoy
m
= S
0
(u
m
) + v
where
S
0
belongs to a setK
(the a priori information on the system) and the noisev
is bounded in somenormY
, i.e.kvk
Y
≤ δ
, for givenδ > 0
. Then itis possible todene the feasiblesystemsetFSS =
{S ∈ K : ky
m
− S(u
m
)
k
Y
≤ δ}
(8)which is the set of systems that are compatible with the measured data and the a priori
assumptions. The set
FSS
can be very complicated, depending on the structure ofK
and the normk · k
Y
. For example, ifK
is the classof LTI systems whose impulse response has anassignedexponentialdecayand noiseis`
∞
-bounded,thenFSS
isaninnitedimensionalpolytope. Obviously,forreal-worldsystems,includingnonlinearitiesortime-varyingdrifts,
FSSmaybemuchmorecomplex.
Sincetheset
FSS
containsalltheinformationprovidedbydataandaprioriassumptions,itisnaturaltoevaluatethequalityofanominalmodel
G
n
according toitsworst-caseerrorwithrespecttoelementsof
FSS
. In otherwords,theidenticationerrorassociatedtoG
n
isgivenby
E(G
n
) =
supS
∈FSS
kS − G
n
k
S
(9)
where
k · k
S
isasuitablenorminthesystemspace.Inorder toidentifyanominalmodel,amodelclassmustbechosen.Acommon
require-ment is that the nominal modelmust be simple(low dimensional,linearly parameterized,
etc.). Hence,atypicalstructurefor
G
n
isG
n
(q; θ) =
n
X
i=1
θ
i
B
i
(q)
(10)wherethe
B
i
(q)
areuserdenedbasisfunctions,suchasFIRlters,LaguerreorKautzlters[23],generalizedorthonormalbasisfunctions[22],etc. Ifwedenoteby
M
thesetof nom-inal models parameterized as in (10), the problem of selecting a model inM
according tooptimalnominalmodelis givenbytheconditionalcentralestimate
G
∗
n
=
arg infG
∈M
supS
∈FSS
kS − Gk
S
.
(11)Since in most practical situation nding an exact solution of the min-maxoptimization
problem(11) isaprohibitivetask,suboptimal estimators areconsidered[5]. Inthispaper,
we will assume that noise is
`
∞
-bounded (i.e.,k · k
Y
= `
∞
) and select as nominal modelG
n
(q, θ
r
)
,whereθ
r
=
arg infθ
∈
IRn
ky
m
−
n
X
i=1
θ
i
B
i
(q)u
m
k
∞
.
(12)This is known as restricted projection estimateand enjoys someniceproperties, including
thefactthatitdoesnotdependontheactualvalueofthenoisebound
δ
. Herethischoiceismotivatedonly by computationalreasons,as
θ
r
canbeeasilycomputedby linear
program-ming;other moresophisticatedestimationalgorithmscanbechosenwithoutmodifyingthe
wholerobustidenticationstrategy.
Whatever nominalmodel hasbeen identied, the actualsize of theunmodeled
dynam-ics can be evaluated from available data by exploiting the Model Error Modeling concept
described in Section3. The identicationof the error system(7) canbeperformed viaset
membership identicationalgorithms, exploiting the noisebound on
v
. If the structure ofthemodelerrormodelischosenas
G
e
(q;
θ) =
n
X
i=1
θ
i
B
i
(q)
(13)thefeasiblesetforthemodelerrorparametersisgivenby
FES =
{
θ
∈
IRn
:
k −
n
X
i=1
θ
i
B
i
(q)u
m
k
Y
≤ δ}.
(14)Then,onemayselecttheworst-caseoptimalerrormodelbycomputingtheChebishevcenter
of
FES
θ
∗
=
arg infθ
∈
IRn
sup ~θ
∈FES
k
θ −
~θ
k
(15)where
k · k
denotestheEuclideannorm(ofcourse,othernorms canbechosen).As for (11), problem (15) maybe computationally infeasible, and suboptimal solutions
are sought. For example, since for
`
∞
-bounded noiseFES
is a polytope in IRn
, it can be
recursivelyapproximatedbysimplerregions,likeellipsoidsorparallelotopes[4,3],andthe
centeroftheseapproximatingsetsmaybechosenasanestimateof
θ
∗
.If robust identicationis orientedto
H
∞
control,anuncertainty bandassociatedtothe nominalmodelfrequencyresponsemustbedeliveredtothecontroldesigner. Inthistheset
FES
ontothecomplexplaneforeachfrequencyofinterest. Thisleadstothefrequencydomainuncertaintyset
V(FES) =
[
ω
V
ω
(FES)
whereV
ω
(FES) =
{z ∈
Cl: z = A
ω
θ,
θ
∈ FES}
(16) andA
ω
= [
B
1
(e
jω
) . . .
B
n
(e
jω
)]
∈
Cl1
×
n
. Once again, the computation of
V(FES)
may be a formidabletaskiftheexactFES
denedby(14)isconsidered. Hence,setapproximationsareusefulalsointhisrespect. Exampleoftheresultingplotsfornominalmodelandmodelerror
willbeshown inSection5.
The overallsetmembershipidenticationstrategycanbesummarizedasfollows.
1. Identify the nominal model
G
n
(q; θ)
, using conditional estimators (12) based on thefeasiblesystemset(8).
2. Computetheresidual
= y
m
− G
n
(q; θ)u
m
.3. Selectamodelerrorstructure,choosethenoisebound
δ
andcompute(orapproximate)thefeasiblemodelerror set(14).
4. Identifyanominalmodelerror
G
e
(q;
θ)
,usingoptimalorsuboptimalestimators based onFES
.5. Mapthe nominalmodelplusthe modelerror and its uncertaintyregiononto the
fre-quencydomain.
Remark. The use of residual data in set membership identication for evaluating the
worst-case
H
∞
norm ofthe unmodeled dynamics has beenintroduced in[9], for standard least-squares nominal models. The above set membership model error modeling strategycanbeseenasageneralframeworkinwhichthestructuresofthenominalanderrormodels,
andthecorrespondingidenticationalgorithmsmustbechosenbytheuseraccordingtothe
specicproblem(a prioriknowledge,noisebound,error norm,etc.).
Inthenextsubsection,itwillbeshownhowitispossibletoexploittheaboveframework
toobtainareliable measureofthe sizeof theunmodeled dynamics. This will alsoprovide
usefulinformationabout theselectionofthenominalmodelclass.
4.2 Model Validationand ModelClass Selection
Lift this to a seperate section (WR+AG)?
In the Set Membership procedure previously described, a key parameter is the noise
Hence,
δ
canbeusedasatuningparameter,inordertoevaluatethe\distancetovalidation"oftheselectednominalmodel.
Letusconsidertheset
V
ω
(FES)
in(16)anddened(ω) =
minz
∈V
ω
(FES)
|z|.
(17) Ifwesetd =
supω
d(ω)
(18)thenitisclearthatthenominalmodelcanbedeemedtobeunfalsiedif
d = 0
. Ifellipsoidalapproximationsof
FES
areconsidered,thesetsV
ω
(FES)
areellipsesinthecomplexplaneand theoptimizationproblem(17) canbeeasilysolved. Then,d
in(18) canbeapproximatedbytakingthemaximumoveranitenumberoffrequenciesintherangeofinterest.
Itisworthobservingthat
d
isafunctionofδ
,sinceFES
(anditsfrequencyimageV
ω
(FES)
) dependon the noisebound. Then,one canlook forthe minimumvalue ofδ
for whichthenominalmodelisstillnotfalsiedbydata,i.e.
δ
∗
=
min{δ: d=0}
δ.
(19)
Since
d
is anincreasingfunctionofδ
, thecomputationofδ
∗
canbeeasilyperformed within
thedesiredprecision, usingastandardbisection on
δ
. Obviously,δ
∗
willnotbelargerthan
thevalueofthenormgivenineqn.(12),whichistheminimumnoisebound forthenominal
modelwhenusingnomodelerror model.
Itisbelievedthatthequantity
δ
∗
isimportantinseveralrespects. Inparticular,itallows
one to deliver the smallest frequency domain uncertainty region associated to the model
error model, which does not falsify the nominal model. This meets a typical requirement
ofthecontroldesigner: the\tightest"uncertaintybandaroundthenominalmodel,without
contradictingtheinformationgivenbydata. Clearly,
δ
∗
dependsonthemodelclassselected
for identicationof the model error model: the\richer" thisclass, thesmaller
δ
∗
. On the
otherhand,atoosimplemodelclassmayresultinalargevalueof
δ
∗
,whichwouldcontradict
the a priori assumptionon the noise size. Hence,
δ
∗
can be usedto \validate" either the
noiseboundor themodelerrormodelclass,accordingtotheavailableaprioriinformation.
Ifareliableaprioriestimate
^
δ
ofthenoisesizeis available,themodelclassformodelerroridenticationshouldbechosensothat
δ
∗
turnouttobeascloseaspossibleto
^
δ
. Conversely,ifthemodelerrormodelclassissuggestedbyaprioriinformationaboutthesystemdynamics,
δ
∗
providesavalidationtoolforthenoisebound(forexample,ifδ
∗
ismuchsmallerthanthe
aprioribound,one mayconcludethatthelatterwasoverestimated).
Anotherusefulpropertyof
δ
∗
isthatitgivesarationaleforselectinganappropriatemodel
classfornominalmodelidentication. Letusdenoteby
p
avectorofparametersthatdenea model class for
G
n
(for example, the numberand/or the pole locations of the family ofmodelandmodelerrormodelassuggestedinSection4.1,andcompute
δ
∗
(p)
. Then,atleast
inprinciple,onecanchoose theoptimal modelstructurebysolving
inf
p
δ
∗
(p).
(20)
In practice, the optimization in (20) canbe performed over a nite set
P = {p
1
. . . p
m
}
(see theexampleinthenextsection,wherep
isthepoleofaLaguerreexpansionofxedorder).5 Simulation Example 5.1 ExperimentalSetup
Gnom
u
zg
zsat
zf
F
y
noiseFigure1: Theexperimentalsetup: Alinearsystem
G
nom
inparallelwithalinearFilterF
andasaturationnonlinearity. Theoutputsignal
y
iscorruptedbynoise.ConsidertheexperimentalsetupasshowninFig.1. Itscoreconsistsofthefollowingfth
orderlinearsystem
G
nom
(s) =
24.15(s + 1)
0.012s
5
+ 0.25s
4
+ 1.80s
3
+ 5.86s
2
+ 7.33s + 1
.
Its dynamics are perturbed by a nonlinearity in the medium frequency range. This is due
toa parallelconnectionwith alinear secondorder lter
F(s) =
2s
0.1s
2
+s+0.09
and a saturation nonlinearity, that saturates at levell = 1.2
. According to theamplitude curve ofthe lter,whichisgiveninFig.2,thenonlinearitywillbe\active"inthefrequencyrange
ω = [0.1; 10]
,dependentontheamplitudeoftheinputsignal. Thesystemwillbehavelinearly,whenthe
outputofthelter
z
f
issmallerthan1.2
inamplitude;theny = z
g
+ z
sat
= (G
nom
+ F)u
holds.In order to be able to apply Stochastic Embedding, the input signal we use is a linear
combination of
53
sinusoids(thecontainedfrequencies aremarkedas crossesinFig. 4). Weuse a sampletime of
T
s
= 0.04s
and3000
samples. We will use four dierent data sets toexaminethebehaviorofthelinearidenticationmethodstothisnonlinear plant,whichare
10
−2
10
−1
10
0
10
1
10
2
10
−1
10
0
10
1
Amplitude
Amplitude, Filter F
Figure2: Amplitudeplotofthesecondorder linearlter
F
.0
20
40
60
80
100
120
−10
−5
0
5
10
input u
0
20
40
60
80
100
120
−40
−20
0
20
40
noisy output y
0
20
40
60
80
100
120
−2
−1
0
1
2
noise
time/s
0
20
40
60
80
100
120
−15
−10
−5
0
5
10
15
filter output z
f
0
20
40
60
80
100
120
−1
−0.5
0
0.5
1
sat output z
sat
time/s
Figure3: Selectedsignalsfromdataset3. Left: Testinput
u
,noisyoutputy
andoutputerrornoise(top-down). Right: Outputofthelter
F
,z
f
,and ofthesaturationnonlinearity,z
sat
.Dataset excitationofsaturation inputlevel(rsthalf) inputlevel(sec.half)
1 none(linearbehavior)
1
1
2 low
1
2
3 medium
1
6
4 high
10
10
Experiments1 and4excitethesystemwith aninputsignal asdescribed abovewithan
am-plitude of about
1
and10
respectively. These input signalswill lead toan either linear orcompletelynonlinearbehavior. Experiments2and3switchfromalowamplitudeinput
sig-naltoahigherone. Wenote thattheresultsobtainedlateron willnotchangesignicantly
when switching form a higher to a lower amplitude. For experiment 3, these signals are
reportedinFig.3.
corre-10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
Amplitude
Spectral analysis of Noiseless Data: D1 (k−), D2 (b:), D3 (g−.), D4 (r−−)
Frequency, Excited frequencies (kx)
10
−2
10
−1
10
0
10
1
10
2
10
−1
10
0
10
1
10
2
Amplitude
Spectral Analysis of Noiseless Data (E1, −); and linear model (−−)
Frequency, Excited frequencies (kx)
Figure4: Left: Spectralanalysis(employingasmallsmoothingwindow)ofthefournoiseless
datasets: dataset1(solid),dataset2(dotted),dataset3(dash-dotted)anddataset4(dashed).
Thecrossesonthefrequencyaxismarkthose53frequencies,containedinthemulti-sinusoid
inputsignal. Right: Bodeplotofthelinear\mode"
G
nom
+ F
(dashed)incomparisontothespectralanalysisofdataset1(solid).
sponding inputsignal
u
. The resultis depictedinFig. 4 (left), togetherwith theBode plotof the linear system
G + F
, which will be \the true plant" for low input level (right). Allfrequency proles are quite similarfor low frequencies, while they are signicantly
dier-ent at higher frequencies. Clearly,the frequency measurements obtained inthis case may
changefor dierent inputsignals, especiallyforaninputsignal with acompletely dierent
powerspectrum. Nevertheless,wekeeptheobtainedfrequencyprolesasareferenceinthe
comparisonofthedierentmethods,forthespecicinputsselected.
Add something about 2nd moment equivalent (LL)
Additionally, theoutput is corruptedby noise, whichis alsoshown in Fig.3 (left). The
noiseisanormallydistributedonewithvariance
0.0102
,addedtoascaledversion(toamax-imum amplitudeof about
1.4
) ofthe200Hz
signal registered during theLomaPrietaearth-quakeintheSantaCruzMountainsinOctober1989(whichhasbeenmeasuredattheCharles
F.RichterSeismologicalLaboratoryandmadebeenavailablebyTheMathWorksInc.). This
noisesignal hasbeen choseninorder toavoid usageof standardstationary stochastic
pro-cessesor boundaryvisitingsignalsasnoisemodels.
Inthefollowing,datasets1-4willbeusedforidenticationpurposes. Dierentdatasets
(obtainedbyaddingadierentoutputnoise)willbeemployedinthemodelvalidationstep,
for those methods that allow an explicit validation procedure (i.e. Model Error Modeling
andSetMembershipIdentication).
Thegoaloftheidenticationexperimentsistoobtainanominalmodelofordernotlarger
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
pole location p
average modeling error SigmaGbar
Figure 5: NSSE: Laguerre pole vs. mean error over all frequencies, in order to choose the
\optimal"Laguerrepole,thatguaranteesminimumerror. Result isbasedondataset4.
sizeforrobustcontrollerdesignusing standard
H
∞
methods.5.2 Identication via Non-Stationary StochasticEmbedding
The rst stepwithin non-stationarystochasticembedding is the estimation ofthe transfer
functionatthosefrequenciescontainedintheinputsignal. Thesecondstepisthenthe
esti-mationofthenominalmodel,basedonthesetransferfunctionpointestimates. Byproblem
denitionwearerestrictedtomodels oforder
4
,thus wechoose suchaLaguerreexpansionforouridentication. Inordertoobtaincomparableresults,weuseacondencelevelof
99
%hereaswellasforthepredictionerrormethodsintheModelErrorModelingapproach. We
remark, however, that theshape of the uncertainty band does not change paramountly at
lowandmediumfrequencies,whendecreasingthecondencelevel. Moreover,preliminary
testrunsshowthattheintegratedrandomwalkwillproduceuncertaintybandsthatarequite
tightforlowerfrequenciesanduselesslargeforhigherfrequencies,whereastherandomwalk
(increasing linearly with frequency)produces reasonableuncertainty bands. Therefore, we
conclude that thesystemdoes not containimportant resonances, and continuewith
error-propagationbyrandomwalkstrategy. WeareleftwiththechoiceofthepoleintheLaguerre
basis. Therefore, weemploythe minimizationofthe average modeling error,as suggested
inSec.2. A typicalplotof meanerror vs. polelocationforxedorder is reportedinFig. 5
(dataset 4).
Choosing pole for the Laguerre expansion to
p = −0.2895
(dataset1,2) andp = −0.5737
(dataset 3,4)respectively, leastsquares estimation ofthe nominalmodel and error
propa-gationleadstotheresults asdepictedinFig. 6. Note, thatchoiceoferror propagation and
condenceleveldonotin uencethenominalmodel. Inallfourcases,theuncertaintyband
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
For Comparison: Spectral analysis of Dataset (b−−)
Frequency
Magnitude
Estimated model
Point Estimate
99.99% conf. cloud, random walk
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
For Comparison: Spectral analysis of Dataset (b−−)
Frequency
Magnitude
Estimated model
Point Estimate
99.99% conf. cloud, random walk
Figure 6: NSSE: Identication based on datasets 3 (left) and 4 (right) respectively. Both
approachesusea
4th
orderLaguerreexpansionwithpolelocatedatp = −0.5737
andrandomwalkaserrorpropagation. The
99
%condencecloudsaredrawn(shaded),aswellastransferfunctionpointestimates(crosses), estimatednominalmodel(solid)andspectralanalysisof
noiselessdata(dashed).
5.3 Identication via Prediction ErrorMethods
WeestimateanOutputErrormodeloforder
4
asnominalmodelusingpredictionerrormeth-ods. The result fordataset1,togetherwiththestandardresidualtest,isreportedinFig.7.
The realplantis clearlynotinthe condenceregion,whichis rathersmalldue tothelarge
amountofdata(3000samples) usedhere. Moreover,the modeldoesnot seemtopass the
classicalresidualtest. A similarstatementholds for the remaining threedatasets. Hence,
weareinneed ofmore detailed informationabout the accuracyof the identied
4rth
oderOE model. Basedon theobtained nominalmodel, wethereforeproceed withestimationof
anerrormodelbasedontheresidualdata(calculatedfromthevalidationdataset). Acrucial
stepistochoosethestructureoftheerror model,sothatthenominalmodelisnotfalsied
byitsownerrormodel,i.e.zerohastobeinsidethe,say99%,condenceregionoftheerror
model.
Wechooseaerror modeloftheform
A(q)(t) =
B(q)
F(q)
u(t − n
k
) +
C(q)
D(q)
v(t)
withpolynomialorders
n
A
= 1, n
B
= 9, n
C
= n
D
= 5, n
F
= 10
anddelayn
k
= 1
forvalidationdataset1. TheresultisreportedinFig.8(left). Themodelerrormodelvalidatesthenominal
model,atthepriceofalargercondenceregion(comparedtotheoneinFig.7).
Wefollowthesamereasoningforvalidationdataset4. Aswefacenon-trivial
straight-10
−2
10
−1
10
0
10
1
10
2
10
−1
10
0
10
1
10
2
Amplitude
Nominal plant (b−) and Real plant (r−−)
0
5
10
15
20
25
−0.2
0
0.2
0.4
0.6
0.8
1
1.2
Correlation function of residuals. Output validation noise data
lag
−25
−20
−15
−10
−5
0
5
10
15
20
25
−0.2
−0.1
0
0.1
0.2
0.3
Cross corr. function between input Sum of 53 sinusoids and residuals from output validation noise data
lag
Figure7: PEM: Left:
4rth
order outputerror model, using dataset1for identication. The99
% condenceregionis givenas well(shaded). Right: Correlation functionoftheresiduals (upperplot) andcross-correlationfunction(lowerplot) usingthevalidation dataset.forward how to choose order and structure of the model error model in order to validate
the nominal model. After several trials, we used an error model with polynomial orders
n
A
= 4, n
B
= 9, n
C
= n
D
= 5, = n
F
= 10
and delayn
k
= 1
, whichdoes not verify thenomi-nalmodel,butgives betterinformationonthefrequencyrangewheretheerror occurs. The
result isreportedinFig.8(right).
5.4 Identication via SetMembershipEstimation
Toobtaina
4rth
ordernominalmodelweusea4rth
orderLaguerreexpansion. ThepolesoftheLaguerreexpansionarechosenaccordingtothestrategysuggestedinSection4.2. Hence
we will use the hard bound on the noise as additional tuning parameter for the selection
ofthe modelclass. A typicalpictureof Laguerrepole locationvs.minimum, non-falsifying
noisebound isreportedinFig.9.
Using dataset 1,3 and 2,4, weend up with a Laguerre pole at
p = 0.97
andp = 0.98
re-spectively. Duetotheincreasingamountofnonlinearity,theminimum,non-falsifyingerror
bounds increaseby
δ = 0.70, 0.83, 1.29, 2.63
fordatasets 1-4 respectively. These \identied"noiseboundsfordatasets1-3coincidequitewellwiththesizeoftheactualnoiseofabout
1
,asgiveninFig.3. Clearly,thehighervaluefordataset4isdue toasignicantcontribution
fromthenonlinearity.
In order toobtainafrequencydomainuncertaintyrepresentation,we followthe
proce-duresuggestedinSection4.2. Toreducethecomputationalburden,weapproximatetheFES
in(14) via ellipsoids[4]. Fig. 10 reports the nal result for datasets1 and 4. Weobserve,
10
0
10
1
10
2
10
−1
10
0
10
1
Nominal plant (b−) and Real plant (r−−)
Magnitude
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Model Error Model With Uncertainty Region
Frequency
Magnitude
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
Nominal plant (b−) and Spectral analysis (r−.)
Magnitude
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Model Error Model With Uncertainty Region
Frequency
Magnitude
Figure8: PEM:Left:
4rth
orderoutputerrormodel,usingdataset1foridentication. Modelerrormodelingwitha
(1,9,5,5,10,1)
-errormodelusingvalidationdataset1. Therealplantis shown as well(dashed). Right:
4rth
order output error model, using dataset4 foriden-tication. Model error modeling with a
(4,9,1,1,11,2)
-errormodel using validation dataset4. Upperplots: nominalmodel(solid)alongwithuncertaintyregion(shaded) and
spec-tralanalysis ofnoiseless data(dashed). Lowerplots: modelerror modelswith uncertainty
regions.
0.9
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
0.6
0.8
1
1.2
1.4
1.6
Laguerre pole
minimum noise bound for model error
Figure9: SMI:LocationofLaguerrepoleversusminimum,non-falsifyingnoisebound ofthe
10
−2
10
−1
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Frequency
Magnitude
Nominal plant (b−) and Real plant (r−−)
10
−2
10
−1
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Frequency
Magnitude
Model error model (b−) with uncertainty region
10
−2
10
−1
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Frequency
Magnitude
Nominal plant (b−) and Spectral analysis (r−.)
10
−2
10
−1
10
0
10
1
10
2
10
−4
10
−2
10
0
10
2
Frequency
Magnitude
Model error model (b−) with uncertainty region
Figure10: SMI:Fordatasets1(left)and4(right): Theupperplotshowsrestrictedprojection
estimate (solid) insidethe unfalsieduncertainty region(shaded) compared to linearplant
(left plot) and spectral analysis of noiseless data (right plot) respectively. The lower plots
showthemodelerror modelinsideits unfalsieduncertaintyregion.
ranges. Moreover,theuncertaintybandisreasonablysmallandincludesthefrequency
pro-lesprovided byspectralanalysis.
5.5
Why the comparison in the next subsection?
(LL)
5.6 Comparison
Thesystemgeneratingthedata(seeFig.1)containsasaturation,whichisanoddstatic
non-linearitywithagainvaryingbetween
0
and1
. Wethereforecomparetheuncertaintyregions,deliveredby identication with the magnitudeenvelope given by
G
nom
and(G
nom
+ F)
aslowerand upperbound respectively. For easeofnotation,wewillrefertothelatterregion
as the \true uncertainty region". Figures 11and 12report the results fordatasets 2 and 4
respectively. The identied uncertainty regions should cover parts of the true uncertainty
region,obviouslydependedoftheexcitationofthenonlinearitybytheinputsignal.
Weobserve,thatforbothdatasetstheuncertaintysetproducedbypredictionerror
meth-odsis misleadinglytight,becauseof alarge amountof data(
1500
samples). Moreover, thenominalmodelisfalsiedforthemediumfrequencyrangefordataset4,seeFig.8.
Moreover, we note that for dataset 4, Fig. 12, preferably the \lower part" of the true
uncertainty region is captured by all methods. This is clear due to the lower gain of the
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Bode Magnitude: nominal model (r−) and G
nom
, G
nom
+F (b−−)
Frequency
Magnitude
Estimated model
Point Estimate
99.99% conf. cloud, random walk
10
−2
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Bode Magnitude: nominal model (b−) and G
nom
,G
nom
+F (b−−)
Magnitude
Frequency
10
−2
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Frequency
Magnitude
Bode Magnitude: nominal model (b−) and G
nom
,G
nom
+F (b−−)
Figure11: Dataset 2: Nominalmodel(solid)anduncertaintyregion(shaded) incomparison
totheamplitudesof
(G
nom
+
1
2
F)
±
1
2
F
(dashed). FromlefttorightforNSSE(continuoustime), MEMandSMI(discretetime).10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Bode Magnitude: nominal model (r−) and G
nom
,G
nom
+F (b−−)
Frequency
Magnitude
Estimated model
Point Estimate
99.99% conf. cloud, random walk
10
−2
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Bode Magnitude: nominal model (b−) and G
nom
,G
nom
+F (b−−)
Magnitude
Frequency
10
−2
10
−1
10
0
10
1
10
2
10
−2
10
−1
10
0
10
1
10
2
Frequency
Magnitude
Bode Magnitude: nominal model (b−) and G
nom
,G
nom
+F (b−−)
Figure12: Dataset 4: Nominalmodel(solid)anduncertaintyregion(shaded) incomparison
totheamplitudesof
(G
nom
+
1
2
F)
±
1
2
F
(dashed). FromlefttorightforNSSE(continuoustime), MEMandSMI(discretetime).Our comparison addresses three dierent approaches to robust identication: Stochastic
Embedding, Model Error Modeling and Set Membership Identication. The rst two
ap-proaches have been developed in the statisticalframework, while the latter relies on UBB
errorassumptions. Duetothedierentnatureofthethreemethods,anoverallcomparison
mightbecomeunfairat acertainstage. Duetorestrictionsonharmonicinputsignal for
ex-ample,onemight notalwaysbeabletoapplyStochasticEmbeddingtoagivendataset. We
thereforeconcentrateourselves onhighlighting themostimportant propertiesand benets
ofeachmethod.
Model Error Modeling and Set Membership Identication allow an explicit validation
step (on freshvalidation data). Asno furtherknowledge than the nominali/o behavior is
needed,theseframeworksalsoallowvalidationofanalreadyexistingmodelfortheprocess.
Evenwhen invalidated, valuable information on the underlying error canbe obtained, for
instanceinwhichfrequencyrangethemodelissuÆcientlyreliable.
Incontrast,StochasticEmbeddingallowsthechoicebetweendierentrandomwalk
mod-els(thevariancemayincreaseindierentwayswithfrequency)toperformtheerror
propa-gation. Althoughthisisnoexplicitvalidationstep,itisaquitereasonableconcepttoproduce
reliablecondenceregions.
The Model Error Modeling setup enjoys exibility in the choiceof the structure of the
nominal model, to end up with for instance OE or ARX models. In contrast, Stochastic
Embeddingand the presentedSetMembership Identicationare(thelatteratleastinthis
context)tiedtoaparameterizationoforthonormalbasisfunctions. Bothschemes,however,
comealongwitharationaleforchoiceofthepoleinaLaguerreexpansion. Itisworthtonote
that both end up with basicallythe samerecommendation(in our caseat
s = −0.29, −0.57
,correspondingto
z = 0.98, 0.97
indiscretetime).Aremainingprobleminthemodelerrormodelingapproachisthecorrectorderand
struc-tureselectionofthemodelerrormodel,whichisingeneralnotstraightforward. This
prob-lemhasalsobeenreportedin[21],wheretheauthorssuggestanadaptiveandnonparametric
frequency-domainmethodthatestimatesthefrequencyresponseofthemodelerrorandalso
allowslocaltuningindierentfrequencybands.
We observe, that all methods deliver an estimated nominal model, along with an
un-certaintyregion,whichis certainlysuitedforrobustcontroller design,forinstanceusingan
H
∞
mixedsensitivity approach, based on the estimateduncertainty band in thefrequency domain.6 Conclusions and Future Works
Wecomparedidenticationmethods,workinginthetimeandfrequencydomainand
dependon the apriori knowledge (for example,if harmonicinput signalsare possibleor a
prioriinformationonthe noiseamplitudeor statistics areavailable). However,weshowed
themain features ofthemethods andobtainedreliablenominalmodels andacceptable
re-lateduncertaintiesinallthreecases.
Some ‘‘nonlinear’’ comments (LL)
Moreover, theideasof modelerror modeling andtheirapplicationinthecontextofset
membership identication have been analyzed. In particular, it has been shown that the
separationofnoiseandunmodeleddynamicsisquitenaturalinthisframework,andthatthe
minimumnoisebound forwhicha nominalmodel is notfalsied by the data canbeeasily
computedand usedasatoolformodelclassselection.
Complementaryworkhasstilltobedone,inseveraldirections. Forexample,more
com-plicatedmodelerrormodelstructuresareneeded,tocopewithnonlinearitiesortime-varying
drifts. Obviously,thisrequiresmoresophisticatedidenticationalgorithmsandsmarter
ap-proximationsinthecomputationoftheuncertaintyregions. Moreover,accuratecriteriafor
the a priori selectionof the structureof the model error model, depending on the specic
control-orientedidenticationproblem,must stillbeinvestigated.
References
[1] J. H. Braslavsky. Model error quantication via non-stationary stochastic
embed-ding: Hairdryer example. CIDAC, Dept of Electrical Engineering, Univ of Newcastle,
Callaghan,Australia.ManualforNSSE-package,Mar.1999.
[2] J.H. Braslavskyand G. C.Goodwin. A noteon non-stationarystochasticembedding
for modelling error quantication in the estimation of resonant systems. Technical
Report EE99014,CIDAC,DeptofElectricalEngineering,UnivofNewcastle,Callaghan,
Australia,Jan.1999.
[3] L. Chisci,A.Garulli, A.Vicino, andG. Zappa. Blockrecursiveparallelotopicbounding
insetmembershipidentication. Automatica,34:15{22,1998.
[4] E.FogelandF.Huang. Onthevalueofinformationinsystemidentication{bounded
noisecase. Automatica,18(12):229{238,Dec.1982.
[5] A. Garulli, B. Z. Kacewicz, A. Vicino, and G. Zappa. Error bounds for conditional
al-gorithmsinrestrictedcomplexitysetmembershipidentication. IEEETrans.on
Proc.oftheSystemIdenticationSymposiumSYSID,pagesWeMD1{3,SantaBarbara,
CA,USA,June2000.
[7] A. Garulli, A. Tesi, and A. Vicino, editors. Robustnessin Identication and Control.
Number 245 in Lecture Notes in Control and Information Sciences. Springer-Verlag,
1999.
[8] A. Garulli,A. Vicino, andG. Zappa. Conditionalcentralalgorithmsfor worst-caseset
membershipidenticationandltering. IEEE Trans.onAutomaticControl,Jan.2000.
[9] L.Giarre, M.Milanese,andM.Taragna.
H
∞
identicationandmodelquality evalua-tion. IEEE Trans.onAutomatic Control,42(2):188{199,1997.[10] G.C.Goodwin. Identicationandrobustcontrol: Bridgingthegap. InProc.ofthe7th
IEEE MediterranianConferenceonControland Automation,Haifa,Israel,June1999.
[11] G.C.Goodwin,J.H.Braslavsky,andM.M.Seron. Non-stationarystochastic
embed-dingfortransferfunctionestimation.InProc.ofthe14thIFACWorldCongress,Bejing,
China,July1999.
[12] G.C.Goodwin,M.Gevers,andB.Ninness.Quantifyingtheerrorinestimatedtransfer
functionswithapplicationtomodelorderselection.IEEETrans.onAutomaticControl,
37(7):913{928,July1992.
[13] R.L.Kosut,G.C.Goodwin,andM.P.Polis. Specialissueon systemidenticationfor
robustcontroldesign. IEEETrans.onAutomaticControl,37(7),July1992.
[14] L. Ljung. Comments on model validation as set membership identication. In
A. Garulli, A. Tesi, and A. Vicino, editors, Robustness in Identication and Control,
pages7{16.Springer,1999.
[15] L. Ljung. Model validation and modelerror modeling. In Proc. of the
Astrom
Sym-posium on Control, pages 15{42, Lund, Sweden, Aug. 1999. Studentliteratur, Lund,
Sweden.
[16] L.Ljung.SystemIdentication{TheoryFortheUser.PTRPrenticeHall,UpperSaddle
River,N.J.,2nd edition,1999.
[17] M. Milanese, J.P. Norton, H. Piet-Lahanier, and E. Walter, editors. Bounding
Ap-proachestoSystemIdentication. PlenumPress,NewYork,1996.
[18] M.Milaneseand A.Vicino. Optimalestimationtheory fordynamicsystemswith
in identication for control. In Proc. of the 38th IEEE Conference on Decision and
Control,pages1488{1493,Phoenix,AZ,USA,Dec.1999.
[20] T.Soderstr om and K.J.
Astrom. Specialissueon trendsinsystemidentication.
Au-tomatica,31(17),Dec.1995.
[21] A. Stenman and F.Tjarnstr om. A nonparametric approach to model error modeling.
InProc oftheIFAC Symposiumon SystemIdenticationSYSID,Jun2000. Accepted.
[22] P. M. J. Van den Hof, P. S. C. Heuberger, and J. Bokor. Systemidentication with
generalizedorthonormalbasisfunctions. Automatica,31(12):1821{1834,1995.
[23] B.Wahlberg. SystemidenticationusingLaguerremodels. IEEE Trans.on Automatic
Control,39:1276{1282,1991.
[24] B. Wahlberg and L. Ljung. Hard frequency-domain model error bounds from
least-squares like identicationtechniques. IEEE Trans. on Automatic Control, 37(7):900{
912,1992.
[25] E.WalterandH.Piet-Lahanier. Estimationofparameterboundsfrombounded-error