Next: 4.5 Existence of optimal
Up: 4.4 Time-optimal control problems
Previous: 4.4.3 Nonlinear systems, singular
Contents
Index
4.4.4 Fuller's problem
Consider again the double
integrator (4.48) with the same control
constraint
. Let the cost functional be
|
(4.66) |
and let the target set be
. In other words,
the final time is free and the final state is the origin, but this
is not a minimal-time problem any more. The Hamiltonian is
and the optimal control must again
satisfy (4.50), i.e., the switching function is
; we can also see this from the general
formula (4.60). The adjoint equation is
|
(4.67) |
We claim that singular arcs are ruled out. Indeed, along a
singular arc
must vanish. Inspecting the differential
equations for
,
, and
, we see that
,
, and
must then vanish too. Thus the only possible
singular arc is the trivial one consisting of the equilibrium at
the origin.
It follows that all optimal controls are bang-bang, with switches
occurring when
equals 0. Our findings up to this point
replicate those in the time-optimal setting of
Section 4.4.1. There, we used the fact that
depended linearly on
to go further and show that the optimal
control has at most one switch. The present adjoint
equation (4.68) is different from (4.49)
and so we can no longer reach the same conclusion. Instead, it turns out that the optimal solution
has the following properties (which we state without proof):
- Optimal controls are bang-bang with infinitely many
switches.
- Switching takes place on the curve
where
.
- Time intervals between
consecutive switches decrease in geometric progression.
The last property is consistent with the fact that the final time
must be finite (as the origin is reachable from every initial
state in finite time, e.g., by using the time-optimal control).
The occurrence of a switching pattern in which switching times
form an infinite sequence accumulating near the final time is
known as Fuller's phenomenon, or Zeno
behavior. We note that the resulting control is measurable but not
piecewise constant (see page ).
Figure 4.17 shows the switching curve4.5 and a sketch of an optimal state trajectory.
Figure:
An optimal trajectory for Fuller's
problem
|
We note that the switching curves in Fuller's problem and in the
time-optimal control problem for the double integrator (treated in
Section 4.4.1) are given by the same formula, but
in the time-optimal problem we had a different value of
, namely,
. The nature of
switching, however, is drastically different in the two cases. We
can embed both problems in the parameterized family of problems
with the cost functional
where
is a parameter. For
we recover the
time-optimal problem while for
we recover Fuller's
problem. Interestingly, one can prove that there exists a
``bifurcation value"
with the following
property: for
the optimal control is bang-bang
with at most one switch, while for
we have Fuller's
phenomenon (Zeno behavior).
The next exercise shows that Fuller's phenomenon is also observed
in time-optimal control problems starting with dimension
. Its
solution can be based on Fuller's problem.
Next: 4.5 Existence of optimal
Up: 4.4 Time-optimal control problems
Previous: 4.4.3 Nonlinear systems, singular
Contents
Index
Daniel
2010-12-20