0% found this document useful (0 votes)
177 views14 pages

Sliding Mode Control

This document discusses sliding mode control, which is a variable structure control method that uses a discontinuous control signal to drive the states of a dynamical system toward a target called the switching surface and maintain the states on this surface. The key points are: 1) Sliding mode control involves designing a switching surface and a control law to drive the system states toward this surface and maintain them on it. 2) The control law has two parts - a corrective control to drive states to the surface, and an equivalent control to maintain motion on the surface. 3) By properly designing the switching surface and control law, the overall system behavior can have desirable properties not present in the individual system structures.

Uploaded by

Asghar Ali
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
177 views14 pages

Sliding Mode Control

This document discusses sliding mode control, which is a variable structure control method that uses a discontinuous control signal to drive the states of a dynamical system toward a target called the switching surface and maintain the states on this surface. The key points are: 1) Sliding mode control involves designing a switching surface and a control law to drive the system states toward this surface and maintain them on it. 2) The control law has two parts - a corrective control to drive states to the surface, and an equivalent control to maintain motion on the surface. 3) By properly designing the switching surface and control law, the overall system behavior can have desirable properties not present in the individual system structures.

Uploaded by

Asghar Ali
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

SLIDINGMODECONTROL

M.SAMIFADALI
PROFESSOREBME
UNIVERSITYOFNEVADA,RENO
1
OUTLINE
Variablestructuresystems
Slidingmodecontrol
Motivatingexample(Khalil)
EquivalentControl
2
VARIABLESTRUCTURESYSTEMS
Dynamicsystemsoftheform
where has discontinuitieswithrespect
tosomearguments
Occurinproblemsinphysics,control
engineeringandmathematics.
Occurnaturallyinsomephysicalsystems,e.g.
forsomeelectricmotorsandpower
converters.
Forsuchsystems,thecontrollawisnaturally
discontinuous
3
VARIABLESTRUCTURECONTROL
Statedependentswitchingfeedbackcontrol
thatintentionallychangesthestructureofthe
system.
Origins:relaycontrol,bangbangcontrol.
Variablestructurecontrolsystem: composedof
independentstructuresandaswitchinglogic.
Overallsystembehaviorisunlikeanyofits
structures.
4
5
EXAMPLE:LINEARPLANT
) ( ) ( ) (
) ( ) (
2 2
2
t u t ax t x
t x t x
+ =
=

1
) ( ) ( t kx t u
1
=
PLANT:
LINEARSTATEFEEDBACK
CONTROLLAW:
Closedloopeigenvalues
2
4
2
1
k a a +
=
2
4
2
1
k a a +
=
For
2
theeigenvalueshavedifferent
propertiesfordifferent and parameter values.
Discontinuous argument

-1 -0.8 -0.6 -0.4 -0.2 0.2 0.4 0.6 0.8
-4
-3
-2
-1
1
2
3
x
2
(t)
x
1
(t)
1
6
Assume a=1and|k|=12
CASE1:
4 12
2
a k > =
4278 . 3 5 . 0
2 , 1
j =
Unstableequilibrium
pointattheorigin.

- 1 - 0.8 - 0. 6 - 0 . 4 -0 . 2 0 .2 0 .4 0.6 0. 8 1
- 4
- 3
- 2
- 1
1
2
3
4
x
1
(t )
x
2
(t)
7
CASE2: 4 12
2
a k < =
, 3
1
= 4
2
=
Saddlepointattheorigin.
Greendashedline(eigenvectorofstableeigenvalue):
systemtrajectoriesconverge totheorigin.
VSSTHEORY: usethisstructure withadiscontinuous to
makethesystemstable.
8
) ( 3 ) (
) ( ) ( ) (
1 2
1 1 2
t x t x
t x t x s
+ =
= x

<
>
=
0 12
0 12
1 1
1 1
x s t x
x s t x
t u
) ( ), (
) ( ), (
) (
x
x
The green dashed line equation candefined as

- 1 - 0 . 8 - 0 . 6 - 0 . 4 -0 . 2 0 . 2 0 . 4 0 . 6 0 . 8 1
- 4
- 3
- 2
- 1
1
2
3
4
x 1(t )
x 2(t )
s(x)=0
Choosethe discontinuouscontrollaw
OVERALLSYSTEMBEHAVIOR
9
Unlikeanyofitsstructures.
VariableStructureSystem(VSS) canpossess
newpropertiesnotpresentinanyofthe
structuresused.
Intheexample,wehave
Case1:UnstableEquilibrium
Case 2:Saddle Point
VSSSystem:Asymptotically Stable
SLIDINGMODES
1
0
switching function ischosen from system
trajectories.
Ingeneral,theswitchingfunctionischosenusingthe system
trajectories.These are known assliding modes.
0 3
1 2
= + = x x s ) (x

-1 -0.5 0.5 1
-3
-2
-1
1
2
3
x
1
(t)
x
2
(t)
x
2
+3x
1
=0
x
2
+c
1
x
1
=0 (0<c
1
<3)
Newswitching
function
SLIDINGMODECONTROL (SMC)
SMCdesign involves twosteps:
(i)Selectionofstablehyperplane(s) inthe
state/errorspaceonwhichmotion should
berestricted,calledtheswitching
function,and
(ii)Synthesisofacontrollaw whichmakes
theselectedslidingsurfaceattractive .
1
1
SMCTRAJECTORIES
Atrajectorystartingfromanonzeroinitial
condition,evolvesintwophases:
a)Reachingmode,inwhichitreachesthe
slidingsurface,and
b)Slidingmode,inwhichthetrajectoryon
reachingtheslidingsurface,remainsthere
foralltimesandthusevolvesaccordingto
thedynamicsspeciedbythesliding
surface.
1
2
1
3
I:DESIGNOF SWITCHINGFUNCTION
PLANTANDSWITCHEDCONTROL
n m nm
Eachmatrixentry:continuouslydifferentiablew.r.t.
Switched(Corrective)Control:
c

SwitchingSurface: dimensionalmanifoldin
n
determinedby constraints.
1 m
1
1
4
1
5
a) Orderofswitchingfunctionislessthanorderofplant
Example:2ndordersystem 1storderswitchingfunction
2 1
b)Slidingmodedoesnotdependonplantdynamicsandis
determinedbyparametersoftheswitchingfunctiononly
(intheexampleon only)
c)Switchingfunctiondoesnotdependonthecontrollaw.

s(x) = u
x
2

x
1

PropertiesofSwitchingFunctions
1
6
Importance ofSwitching Functions
c
2
c
1
c
3
c
1
<c
2
<c
3
e
I. II.
III. IV.
e
1
(0),(0)
2
e
2
e
e
1
UpperLimitc
3
: dependsonthephysicalpropertiesofthe
systemandthetechnologyused.
LowerLimitc
1
: dependsontheallowabletrackingtime
TradeoffbetweenPERFORMANCEandROBUSTNESS
) (
2 2
2 1
s ksign ce e
e e
=
=

1
7
Switching Function Design
Linear
e
e(0),(0)
e
e
Nonlinear
e(0),(0)
e
e
e
ADVANTAGES
+Appropriateforglobaldynamic
propertiesofnonlinearsystems
+Numerousdesignoptions
DISADVANTAGES
Difficulttofindnonlinear functions
Difficulttoobtainthesurface
parameters
ADVANTAGES
+Easytoobtainthesurfaceparameters
DISADVANTAGES
Maynotbeappopriateforsystem
dynamics, ingeneral.
Magnitudeofthecontrolsignalincreases
directlyproportionaltothetrackingerror.
Fewer designoptions
1
8
TimeVarying
Switching Function Design
e
e(0),(0)
e
e
Constant
e(0),(0)
e
e
e
e
e(0),(0)
e
e
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
e1
e2
1
9
Switching Function Design
MostcommonchoiceisLTI SwitchingSurfaces
mn
n m
e
e
n = 2, m = 1
x
0
x
0
s
1
=0
s
2
=0
s=0
n = S, m = 2
2
0
II:FINDACONTROLLAWTOREACH
ANDSTAYTHEREAFTER
u(t)=u
c
(t)+u
eq
(t)
CORRECTIVECONTROL
(compensatethedeviations
fromthe sliding surface
to reach the sliding surface)
EQUIVALENTCONTROL
(makesthederivative
oftheslidingsurface
equalzero tostayon
theslidingsurface)
AttractiveSurface(SlidingSurface):Trajectories
outsidethesurfaceconvergetoit.Onceonthe
surface,trajectoriesremainonit
2
1
EXISTENCEOF SLIDING MODE
Asystemwith inputscanhave
switchingfunctionsand
m
sliding
surfaces.
Thecontrollawdesignandexistenceof
slidingmodearesurveyedin:
Hunget.al.,VariableStructureControl.A
Survey,IEEETran.IndustrialElectronics,40(1),
1993.
DirectSwitchingApproach
ReachingLawApproach
LyapunovFunctionApproach
LYAPUNOVFUNCTIONAPPROACH
PositivedefiniteLyapunovfunction
1
Derivative
v x
x
1
1
Choose foranegativedefinite :
trajectoriesconvergetothesurface.
2
2
SWITCHINGSURFACE
Correctivecontrol
c
:Usehighspeed
switchingtodrivethestatetrajectorytoa
specifiedswitchingsurface.
Attractivesurface:trajectoriesoutsidethe
surfaceconvergetoitandonesstartingonthe
surfacestayonit.
Localcontrol
I
:Designsurfacesothatthe
systemhasgooddynamicbehavioron
(e.g.bypoleplacementforalinear
surface).
2
3
CORRECTIVECONTROL
1
Choosetheinputamplitudesufficientlylarge
tomake negativedefiniteforany :
Robustw.r.t.modelingerrors.
Typicalchoice
c

mm
2
4
CORRECTIVECONTROL :
SISYSTEM/SCALARLINEARSURFACE
mn
1
CorrectiveControl:
c

2
5
EQUIVALENTCONTROL
Motiontangenttotheswitchingsurface
cq
-1
Equivalentdynamics(onthesurface)
-1
2
6
EQUIVALENTCONTROL :
LINEARSYSTEM/LINEARSURFACE
mn
Motiontangenttotheswitchingsurface
EquivalentControl:
eq
-1
SlidingModeDynamics(equivalentsystem):order
zeroeigenvalues
-1
Constraint:
2
7
POLEPLACEMENTDESIGN
LinearDynamics
Similaritytransformation
-1
1 2 1
n-m
2
m
1 11 1 12 2
2 21 1 22 2 2
2
mm
nonsingular
2
8
LEMMA:CONTROLLABILITY
If iscontrollablethen
11 12
is
controllable.
Proof:Controllabilityisinvariantunder
similaritytransformation,
11 12
21 22 2
2
11 12
2
9
LEMMA:CONTROLLABILITY
If iscontrollablethen
11 12
is
controllable.
Proof:Controllabilityisinvariantunder
similaritytransformation,
11 12
21 22 2
2
11 12
3
0
EQUIVALENTCONTROL
cq
-1
eq
-1
Usethetransformedsystem
2
1 2
11 12
21 22
eq 2 2
-1
1 11 2 21 1
1 12 2 22 2
2
-1 2
-1
1 11 21 1
2
-1
1 12 22 2
3
1
ONSWITCHINGSURFACE
2
1 2
11 12
21 22
eq 2
-1 2
-1
1 11 21 1
2
-1
1 12 22 2
1 1 2 2 2 2
-1
1 1
1 11 1 12 2 11 12 2
-1
1 1
Assign eigenvaluesusingpoleplacementtoselect
2
-1
1
andselectanynonsingular
2 2
Eigenvaluesinvariantundersimilaritytransformation
eq 2 2
-1
1 11 2 21 1 12 2 22
-1
3
2
EXAMPLE
z z
11 12
21 22
3
3
MATLAB
>>k=place(A11,A12,5)
k=
1.6000
3.2000
>>Ku=[k*A11+A21,k*A12+A22]/T
Ku=
1.20002.60001.4000
9.60003.20006.2000
>>eig(AB*Ku)
ans =
0.0000
5.0000
0
3
4
CHATTERING
Intheory,thetrajectoriesslidealongthe
switchingfunction.
Inpractice,thereishigh frequencyswitching.
Occursinthevicinityoftheswitchingsurface
duetononidealswitchinge.g.delays,
hysteresis,etc.
Calledchatteringbecauseofthesoundmade
byoldmechanicalswitches.
3
5
EXAMPLE(SLOTINE)
1 2
2
ModelUncertainty
2 1 1
2 1 1 1 2
1 2
3
6
SLIDINGCONDITION
1 2
1 2
Choose
3
7
EXAMPLE(KHALIL)
1 2
2
unknown
0
2
Designastatefeedbacklawthat
stabilizestheorigin.
3
8
SLIDINGMANIFOLD
Constrainthesystemtothesurface
(manifold)
1 2
Motiononmanifold
1 2 1
Stablefor forany .
Drivethetrajectoriestothesurfaceand
maintainthemonit.
3
9
LYAPUNOVFUNCTION
2
1 1 2
1 2
2
2

2
4
0
SLIDINGMODECONTROL
Assume
2
2
2
0 0
>0
0 0 0
4
1
COMPARISONPRINCIPLE
0 0
satisfies
+
0 0
0 0
Decreasestillitreachesitsminimum( )
Themanifold isreachedinfinitetime.
Staystherebecause
0 0
4
2
CONSTANTBOUND
Assume
1
,i.e.
2
1
2
1
(relay)
Relayleadstoafiniteregionofattraction
(invariantset)
4
3
REGIONOFATTRACTION
(invariantset)
1 2 1
1 1
2
1 1 1 1
2
1
1
2
1 1
1 1
Invariantset
1
c
u
4
4
TRAJECTORIESIN
Trajectoriesapproachtheinvariantset
1
c
u
Insidetheinvariantset,trajectories
approach if
2
1
4
5
REGIONOFATTRACTION
4
6
5 0 5
-4
-3
-2
-1
0
1
2
3
4
-5 -3 -1 1 3 5
SPECIFICSYSTEM:PENDULUM
1 2
2 1 2
1 2
2
2 1 2
2 1 1
Let
2 1
Simulatethesystemassumingnegligibleswitching
delays.
4
7
SIMULATIONDIAGRAM
4
8
NOSWITCHINGDELAYS
4
9
0 0.5 1 1.5 2
-2.5
-2
-1.5
-1
-0.5
0
SWITCHINGDELAYS
Chattering:highfrequencyswitching.
Occursinthevicinityoftheswitchingsurface
duetoswitchingdelays.
Iftherelayswitchestopositivefromnegativeat
andfromnegativetopositiveat ,
thesystemexhibitschatteringbehavior.
5
0
CHATTERING(SWITCH
5
1
0 0.2 0.4 0.6 0.8 1
-1.6
-1.4
-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
REDUCINGCHATTERING
Addacontinuouscontrolcomponent
Changethesgn(.)functiontosat(.):linear
controlinsideaboundarylayerofwidth

5
2
EXAMPLE
Addacontinuouscontrolcomponent
Changethesgn(.)functiontosat(.):linear
controlinsideaboundarylayerofwidth
5
3
SWITCHINGDELAY0.1
5
4
-0.2 0 0.2 0.4 0.6 0.8 1
-1.6
-1.4
-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
Ultimatelybounded
butdoesnotconverge
totheorigin.
0 0.2 0.4 0.6 0.8 1
-1.6
-1.4
-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
With signum function
5
5
SlidingModeControlis
DETERMINISTIC(only bounds ofvariations are
considered)
NONLINEAR(the corrective term isnonlinear)
ROBUST (onceontheslidingsurface,thesystemis
robusttoBOUNDEDPARAMETERSVARIATIONSand
BOUNDEDDISTURBANCES)
Switchingfunctiondesignandcontrollawdesign
determinetheperformance.
Althoughchatteringisaproblem,variousmethods
areavailableformitigatingoreliminatingit.
CONCLUSION
REFERENCES
1. R.A.DeCarlo,S.H.Zak,andG.P.Mathews,Variable
StructureControlofNonlinearMultivariableSystems:A
Tutorial,ProceedingsIEEE,Vol.76,No.3,March1988.
2. H.Khalil,NonlinearSystems,PrenticeHall,UpperSaddle
River,NJ,2003.
3. J.J.Slotine andW.Li,AppliedNonlinearControl,Prentice
Hall,EnglewoodCliffs,NJ,1991.
4. KarKeungYoung,P.Kokotovic,andV.Utkin,Asingular
perturbationanalysisofhighgainfeedbacksystems,IEEE
Trans.Automat.Contr.,Vol.22,No.6,pp931938,1977.
5. S.H.Zak,SystemsandControl,OxfordUniv.Press,NY,2003.
5
6

You might also like