104
CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
8.3 Variable Final Time
8.3.1 Main result.
In the problem considered up to now the final time t
f
is assumed to be fixed. In many important cases the final time is itself a decision variable. One such case is the minimum-time problem where
we want to transfer the state of the system from a given initial state to a specified final state in minimum time. More generally, consider the optimal control problem 8.27.
Maximize Z
t
f
t
f t, xt, utdt
subject to dynamics:
˙xt = f t, x, t, ut, , t ≤ t ≤ t
f
, initial condition:
g xt
= b ,
final condition: g
f
xtf = b
f
, control constraint:
u· ∈ U , final-time constraint:
t
f
∈ t , ∞ .
8.27
We analyze 8.27 by converting the variable time interval [t
, t
f
] into a fixed-time interval [0, 1]. This change of time-scale is achieved by regarding
t as a new state variable and selecting a new time variable
s which ranges over [0, 1]. The equation for t is
dts ds
= αs , 0 ≤ s ≤ 1 , with initial condition
t0 = t .
Here αs is a new control variable constrained by αs ∈ 0, ∞. Now if x· is the solution of
˙xt = f t, xt, ut , t ≤ t ≤ t
f
, xt = x
8.28 and if we define
zs = xts, vs = uts , 0 ≤ s ≤ 1 , then it is easy to see that
z· is the solution of
dz ds
s = αsf s, zs, vs , 0 ≤ s ≤ 1 z0 = x .
8.29 Conversely from the solution
z· of 8.29 we can obtain the solution x· of 8.28 by xt = zst , t
≤ t ≤ t
f
, where
s· : [t , t
f
] → [0, 1] is the functional inverse of st; in fact, s· is the solution of the differential equation
˙st = 1αst, st = 0.
8.3.
VARIABLE FINAL TIME 105
With these ideas in mind it is natural to consider the fixed-final-time optimal control problem 8.30, where the state vector
t, z ∈ R
1+m
, and the control α, v ∈ R
1+p
: Maximize
Z
1
f ts, zs, vsαsds
subject to dynamics:
˙zs, ˙ts = f ts, zs, vsαs, αs, initial constraint:
g z0 = b
, t0 = t ,
final constraint: g
f
z1 = b
f
, t1 ∈ R , control constraint:
vs, αs ∈ Ω × 0, ∞ for
0 ≤ s ≤ 1 and v·, α· piecewise continuous. 8.30
The relation between problems 8.27 and 8.30 is established in the following result. Lemma 1:
i Let x
∗
∈ T , u
∗
· ∈ U, t
∗ f
∈ t , ∞ and let x
∗
t = φt, t , x
∗
, u
∗
· be the corresponding trajectory. Suppose that
x
∗
t
∗ f
∈ T
f
, and suppose that u
∗
·, x
∗
, t
∗ f
is optimal for 8.27. Define
z
∗
, v
∗
·, and α
∗
· by z
∗
= x
∗
v
∗
s = u
∗
t + st
∗ f
− t α
∗
s = t
∗ f
− t , 0 ≤ s ≤ 1 ,
, 0 ≤ s ≤ 1 . Then
v
∗
·, α
∗
·, z
∗
is optimal for 8.30. ii Let
z
∗
∈ T , and let
v
∗
·, α
∗
· be an admissible control for 8.30 such that the correspond- ing trajectory
t
∗
·, z
∗
· satisfies the final conditions of 8.30. Suppose that v
∗
·, α
∗
·, z
∗
is optimal for 8.30. Define x
∗
, u
∗
· ∈ U, and t
∗ f
by x
∗
= z
∗
, u
∗
t = v
∗
s
∗
t , t ≤ t ≤ t
∗ f
, t
∗ f
= t
∗
1 , where
s
∗
· is functional inverse of t
∗
·. Then u
∗
·, z
∗
, t
∗ f
is optimal for 8.27.
Exercise 1: Prove Lemma 1.
Theorem 1: Let
u
∗
· ∈ U, let x
∗
∈ T , let
t
∗ f
∈ 0, ∞, and let x
∗
t = φt, t , x
∗
, u
∗
·, t ≤ t ≤ t
f
, and suppose that x
∗
t
∗ f
∈ T
f
. If u
∗
·, x
∗
, t
∗ f
is optimal for 8.27, then there exists a function
˜ p
∗
= p
∗
, p
∗
: [t , t
∗ f
] → R
1+m
, not identically zero, and with
p
∗
t ≡ constant and p
∗
t ≥ 0, satisfying augmented adjoint equation:
·
˜ p
∗
t = −[
∂ ˜ f
∂ ˜ x
t, x
∗
t, u
∗
t]
′
˜ p
∗
t , 8.31
initial condition: p
∗
t ⊥T
x
∗
, 8.32
106
CHAPTER 8. CONINUOUS-TIME OPTIMAL CONTROL
final condition: p
∗
t
∗ f
⊥T
f
x
∗
t
∗ f
. 8.33
Also the maximum principle ˜
Ht, x
∗
t, ˜ p
∗
t, u
∗
t = ˜ M t, x
∗
t, ˜ p
∗
t , 8.34
holds for all t ∈ [t
, t
f
] except possibly for a finite set. Furthermore, t
∗ f
must be such that ˆ
Ht
∗ f
, x
∗
t
∗ f
, ˜ p
∗
t
∗ f
, u
∗
t
∗ f
= 0 . 8.35
Finally, if f
and f do not explicitly depend on t, then ˆ
M t, x
∗
t, ˜ p
∗
t ≡ 0. Proof:
By Lemma 1, z
∗
= x
∗
, v
∗
s = u
∗
t + st
∗ f
− t and α
∗
s = t
∗ f
− t for 0 ≤ s ≤ 1
constitute an optimal solution for 8.30. The resulting trajectory is z
∗
s = x
∗
t + st
∗ f
− t , t
∗
s = t + st
∗ f
− t , 0 ≤ s ≤ 1 , so that in particular
z
∗
1 = x
∗
t
∗ f
. By Theorem 1 of Section 2, there exists a function ˜
λ
∗
= λ
∗
, λ
∗
, λ
∗ n+1
: [0, 1] → R
1+n+1
, not identically zero, and with
λ
∗
s ≡ constant and λ
∗
s ≥ 0, satisfying
adjoint equation:
˙λ
∗
t ˙λ
∗
t ˙λ
∗ n+1
t
= −
{[
∂f ∂z
t
∗
s, z
∗
s, v
∗
s]
′
λ
∗
s +[
∂f ∂z
t
∗
s, z
∗
s, v
∗
s]
′
λ
∗
s}α
∗
s {[
∂f ∂t
t
∗
s, z
∗
s, v
∗
s]
′
λ
∗
s +[
∂f ∂t
t
∗
s, z
∗
s, v
∗
s]
′
λ
∗
s}α
∗
s
8.36
initial condition: λ
∗
0⊥T z
∗
8.37 final condition:
λ
∗
1⊥T
f
z
∗
1 , λ
∗ n+1
1 = 0 . 8.38
Furthermore, the maximum principle λ
∗
sf t
∗
s, z
∗
s, v
∗
sα
∗
s +λ
∗
s
′
f t
∗
s, z
∗
s, v
∗
sα
∗
s + λ
∗ n+1
sα
∗
s = sup{[λ
∗
sf t
∗
s, z
∗
s, wβ +λ
∗
s
′
f t
∗
s, z
∗
s, wβ + λ
∗ n+1
sβ]|w ∈ Ω, β ∈ 0, ∞} 8.39
holds for all s ∈ [0, 1] except possibly for a finite set.
Let s
∗
t = t − t t
∗ f
− t , t
≤ t ≤ t
∗ f
, and define ˜
p
∗
= p
∗
, p
∗
: [t , t
∗ f
] → R
1+n
by p
∗
t = λ
∗
s
∗
t, p
∗
t = λ
∗
s
∗
t, t ≤ t ≤ t
∗ f
. 8.40
First of all, ˜
p
∗
is not identically zero. Because if ˜
p
∗
≡ 0, then from 8.40 we have λ
∗
, λ
∗
≡ 0 and then from 8.36,
λ
∗ n+1
≡ constant, but from 8.38, λ
∗ n+1
1 = 0 so that we would have ˜ λ
∗
≡ 0
8.3.
VARIABLE FINAL TIME 107
which is a contradiction. It is trivial to verify that ˜
p
∗
· satisfies 8.31, and, on the other hand 8.37 and 8.38 respectively imply 8.32 and 8.33. Next, 8.39 is equivalent to
λ
∗
sf t
∗
s, z
∗
s, v
∗
s +λ
∗
s
′
f t
∗
s, z
∗
s, v
∗
s + λ
∗ n+1
s = 0 8.41
and λ
∗
sf t
∗
s, z
∗
s, v
∗
s + λ
∗
s
′
f t
∗
s, z
∗
s, v
∗
s = Sup {[λ
∗
sf t
∗
s, z
∗
s, w + λ
∗
s
′
f t
∗
s, z
∗
s, w]|w ∈ Ω}. 8.42
Evidently 8.42 is equivalent to 8.34 and 8.35 follows from 8.41 and the fact that λ
∗ n+1
1 = 0. Finally, the last assertion of the Theorem follows from 8.35 and the fact that ˜
M t, x
∗
t, ˜ p
∗
t ≡ constant if
f , f are not explicitly dependent on t.
♦
8.3.2 Minimum-time problems