<s>
Automated	B-Application
planning	I-Application
and	I-Application
scheduling	I-Application
,	O
sometimes	O
denoted	O
as	O
simply	O
AI	B-Application
planning	I-Application
,	O
is	O
a	O
branch	O
of	O
artificial	B-Application
intelligence	I-Application
that	O
concerns	O
the	O
realization	O
of	O
strategies	O
or	O
action	O
sequences	O
,	O
typically	O
for	O
execution	O
by	O
intelligent	B-General_Concept
agents	I-General_Concept
,	O
autonomous	O
robots	O
and	O
unmanned	O
vehicles	O
.	O
</s>
<s>
Unlike	O
classical	O
control	O
and	O
classification	B-General_Concept
problems	O
,	O
the	O
solutions	O
are	O
complex	O
and	O
must	O
be	O
discovered	O
and	O
optimized	O
in	O
multidimensional	O
space	O
.	O
</s>
<s>
Solutions	O
usually	O
resort	O
to	O
iterative	O
trial	O
and	O
error	O
processes	O
commonly	O
seen	O
in	O
artificial	B-Application
intelligence	I-Application
.	I-Application
</s>
<s>
These	O
include	O
dynamic	B-Algorithm
programming	I-Algorithm
,	O
reinforcement	O
learning	O
and	O
combinatorial	O
optimization	O
.	O
</s>
<s>
Languages	O
used	O
to	O
describe	O
planning	B-Application
and	I-Application
scheduling	I-Application
are	O
often	O
called	O
action	B-Application
languages	I-Application
.	O
</s>
<s>
In	O
AI	B-Application
planning	I-Application
,	O
planners	O
typically	O
input	O
a	O
domain	O
model	O
(	O
a	O
description	O
of	O
a	O
set	O
of	O
possible	O
actions	O
which	O
model	O
the	O
domain	O
)	O
as	O
well	O
as	O
the	O
specific	O
problem	O
to	O
be	O
solved	O
specified	O
by	O
the	O
initial	O
state	O
and	O
goal	O
,	O
in	O
contrast	O
to	O
those	O
in	O
which	O
there	O
is	O
no	O
input	O
domain	O
specified	O
.	O
</s>
<s>
The	O
most	O
commonly	O
used	O
languages	O
for	O
representing	O
planning	O
domains	O
and	O
specific	O
planning	O
problems	O
,	O
such	O
as	O
STRIPS	B-Application
and	O
PDDL	B-Application
for	O
Classical	O
Planning	O
,	O
are	O
based	O
on	O
state	O
variables	O
.	O
</s>
<s>
Since	O
a	O
set	O
of	O
state	O
variables	O
induce	O
a	O
state	O
space	O
that	O
has	O
a	O
size	O
that	O
is	O
exponential	O
in	O
the	O
set	O
,	O
planning	O
,	O
similarly	O
to	O
many	O
other	O
computational	O
problems	O
,	O
suffers	O
from	O
the	O
curse	B-Algorithm
of	I-Algorithm
dimensionality	I-Algorithm
and	O
the	O
combinatorial	O
explosion	O
.	O
</s>
<s>
An	O
alternative	O
language	O
for	O
describing	O
planning	O
problems	O
is	O
that	O
of	O
hierarchical	B-Application
task	I-Application
networks	I-Application
,	O
in	O
which	O
a	O
set	O
of	O
tasks	O
is	O
given	O
,	O
and	O
each	O
task	O
can	O
be	O
either	O
realized	O
by	O
a	O
primitive	O
action	O
or	O
decomposed	O
into	O
a	O
set	O
of	O
other	O
tasks	O
.	O
</s>
<s>
reduction	O
to	O
the	O
propositional	B-Algorithm
satisfiability	I-Algorithm
problem	I-Algorithm
(	O
satplan	B-Application
)	O
.	O
</s>
<s>
reduction	O
to	O
Model	B-Application
checking	I-Application
-	O
both	O
are	O
essentially	O
problems	O
of	O
traversing	O
state	O
spaces	O
,	O
and	O
the	O
classical	O
planning	O
problem	O
corresponds	O
to	O
a	O
subclass	O
of	O
model	B-Application
checking	I-Application
problems	O
.	O
</s>
<s>
Temporal	O
planning	O
is	O
closely	O
related	O
to	O
scheduling	B-Application
problems	O
when	O
uncertainty	O
is	O
involved	O
and	O
can	O
also	O
be	O
understood	O
in	O
terms	O
of	O
timed	B-Application
automata	I-Application
.	O
</s>
<s>
The	O
Simple	O
Temporal	O
Network	O
with	O
Uncertainty	O
(	O
STNU	O
)	O
is	O
a	O
scheduling	B-Application
problem	O
which	O
involves	O
controllable	O
actions	O
,	O
uncertain	O
events	O
and	O
temporal	O
constraints	O
.	O
</s>
<s>
Dynamic	O
Controllability	O
for	O
such	O
problems	O
is	O
a	O
type	O
of	O
scheduling	B-Application
which	O
requires	O
a	O
temporal	O
planning	O
strategy	O
to	O
activate	O
controllable	O
actions	O
reactively	O
as	O
uncertain	O
events	O
are	O
observed	O
so	O
that	O
all	O
constraints	O
are	O
guaranteed	O
to	O
be	O
satisfied	O
.	O
</s>
<s>
Deterministic	O
planning	O
was	O
introduced	O
with	O
the	O
STRIPS	B-Application
planning	O
system	O
,	O
which	O
is	O
a	O
hierarchical	O
planner	O
.	O
</s>
<s>
Hierarchical	O
planning	O
can	O
be	O
compared	O
with	O
an	O
automatic	O
generated	O
behavior	B-Application
tree	I-Application
.	O
</s>
<s>
The	O
disadvantage	O
is	O
,	O
that	O
a	O
normal	O
behavior	B-Application
tree	I-Application
is	O
not	O
so	O
expressive	O
like	O
a	O
computer	O
program	O
.	O
</s>
<s>
Conditional	O
planning	O
overcomes	O
the	O
bottleneck	O
and	O
introduces	O
an	O
elaborated	O
notation	O
which	O
is	O
similar	O
to	O
a	O
control	O
flow	O
,	O
known	O
from	O
other	O
programming	O
languages	O
like	O
Pascal	B-Application
.	O
</s>
<s>
It	O
is	O
very	O
similar	O
to	O
program	B-Application
synthesis	I-Application
,	O
which	O
means	O
a	O
planner	O
generates	O
sourcecode	O
which	O
can	O
be	O
executed	O
by	O
an	O
interpreter	O
.	O
</s>
<s>
It	O
has	O
to	O
do	O
with	O
uncertainty	O
at	B-Library
runtime	I-Library
of	O
a	O
plan	O
.	O
</s>
<s>
For	O
example	O
,	O
if	O
an	O
object	O
was	O
detected	O
,	O
then	O
action	O
A	B-Application
is	I-Application
executed	O
,	O
if	O
an	O
object	O
is	O
missing	O
,	O
then	O
action	O
B	O
is	O
executed	O
.	O
</s>
<s>
A	O
major	O
advantage	O
of	O
conditional	O
planning	O
is	O
the	O
ability	O
to	O
handle	O
partial	B-Application
plans	I-Application
.	O
</s>
<s>
An	O
agent	O
is	O
not	O
forced	O
to	O
plan	O
everything	O
from	O
start	O
to	O
finish	O
but	O
can	O
divide	O
the	O
problem	O
into	O
chunks	B-General_Concept
.	O
</s>
<s>
For	O
a	O
contingent	O
planning	O
problem	O
,	O
a	O
plan	O
is	O
no	O
longer	O
a	O
sequence	O
of	O
actions	O
but	O
a	O
decision	B-Algorithm
tree	I-Algorithm
because	O
each	O
step	O
of	O
the	O
plan	O
is	O
represented	O
by	O
a	O
set	O
of	O
states	O
rather	O
than	O
a	O
single	O
perfectly	O
observable	O
state	O
,	O
as	O
in	O
the	O
case	O
of	O
classical	O
planning	O
.	O
</s>
