Datasets:
tensorkelechi
/
arxiv-cstext

Dataset card Data Studio Files Community
1
text
stringlengths
0
89.3k
subject to constraints 3and in closedloop with 33
is inputtostate stable ISS in Xfeasfor any admissible
sequence of combined disturbances η where Xfeasis the
set of all initial states xkfor which 32is feasible
PROOF Due to Theorem 3 we know that 32 is re
cursively feasible for all xk Xfeas Therefore we can
prove ISS by showing that the optimal value function
of 32 is an ISSLyapunov function in Xfeasaccord
ing to Definition 5 3 Lemma B44 Denote 32a as
VNxkzv and the optimal value function of 32
asV
Nxk ie VN evaluated at the optimal solu
tionzv Using the candidate sequences ˆzˆvdefined
in 37 it follows that
V
Nxk1V
NxkVNxk1 ˆzˆvV
Nxk
lfˆzNlfz
N
N1X
i0lˆziˆvilz
i v
i
10We then use the candidate sequences 37 and uniform
continuity of lf with corresponding Kfunction βf
to obtain the upper bound
lfˆzNlfz
N lfAKfz
N Γwklfz
N
lfAKfz
NlfAKfz
N
lfAKfz
Nlfz
N
lfAKfz
NΓwklfAKfz
N
lfAKfz
Nlfz
NβfΓwk
We rewrite the sum of the cost differences as
N1X
i0lˆziˆvilz
i v
i lz
0 v
0 lˆzN1ˆvN1
N1X
i1lˆzi1ˆvi1lz
i v
i
and use the candidate sequences 37 and uniform conti
nuity of l with βx βu the respective Kfunctions
to obtain
lˆzi1ˆvi1lz
i v
ilz
iΦe
i0wk v
iΦν
i0wk
lz
i v
i
lz
iΦe
i0wk v
iΦν
i0wk
lz
i v
iΦν
i0wk
lz
i v
iΦν
i0wklz
i v
i
βxΦe
i0wkβuΦν
i0wk
Additionally we rewrite
lˆzN1ˆvN1 lz
NΦe
N0wk Kfz
NΦν
N0wk
lz
NΦe
N0wk Kfz
NΦν
N0wk
lz
N Kfz
NΦν
N0wk