text
stringlengths
0
89.3k
N0Ξ
0
wk
d
Kfz
N Ψd
N0wk
ˆΨd
N1j d
AˆΦe
N1j d
BˆΦν
N1jˆΣNj
d
AΦe
Nj1 d
BΦν
Nj1Ξ
j1 Ψd
Nj1
ˆΨd
Nj d
AˆΦe
Nj d
BˆΦν
NjˆΞj
d
AΦe
Nj d
BΦν
NjΞ
j Ψd
Nj
Using these candidates we then check the inclu
sions 32k and 32l ie for i 0 N2 we get
ˆψd
i i1M
j0ˆΨd
ijW W
ψd
i1 Ψd
i10wk iM
j1Ψd
i1jW W
ψd
i1 iM
j0Ψd
i1jW W32k
σ
i2W ˆσi1W
and for iN1 we get
ˆψd
N1 N2M
j0ˆΨd
N1jW W
d
Kfz
N Ψd
N0wk N1M
j1Ψd
NjW W
αd
KfZfN1M
j0Ψd
NjW W32l
σ
NW ˆσNW
both of which satisfy 32k Finally 32l is trivially fulfilled by the candidate solution ie
ˆαd
KfZfN1M
j0ˆΨd
NjW W
αd
KfZfN1M
j0Ψd
NjW W32l
σ
NW ˆσNW
Since we impose 32k and 32l for each vertex Dd
individually the argument holds for all vertices of D
Therefore we have shown that the candidate solu
tions 37 38 satisfy all constraints in 32 which
proves recursive feasibility 2
Next we show that the closedloop system is inputto
state stable ISS where ISS is formally defined in 3
Definition B42 For the ISS proof we rely on an ISS
Lyapunov function 3 Appendix B6 which we define in
the following definition Note that due to the equivalence
of 1 and 35 we only need to check the stability of 35
to conclude ISS of the original system
Definition 5 ISSLyapunov function Consider
an RPI set S Rnwith the origin in its interior A
function VS R0is called an ISSLyapunov func
tion in Sfor system 35withuκMPCxandwW
if there exist Kfunctions γ1 γ2 γ3 and γ4 such that for
allx SandwW
γ1xVxγ2x 40a
VxVxγ3xγ4w 40b
where is any vector norm on Rn
Theorem 4 Given that Assumption 1 holds system 1