<s>
Simultaneous	B-Application
localization	I-Application
and	I-Application
mapping	I-Application
(	O
SLAM	O
)	O
is	O
the	O
computational	O
problem	O
of	O
constructing	O
or	O
updating	O
a	O
map	O
of	O
an	O
unknown	O
environment	O
while	O
simultaneously	O
keeping	O
track	O
of	O
an	O
agent	B-General_Concept
's	O
location	O
within	O
it	O
.	O
</s>
<s>
Popular	O
approximate	O
solution	O
methods	O
include	O
the	O
particle	B-Algorithm
filter	I-Algorithm
,	O
extended	O
Kalman	O
filter	O
,	O
covariance	O
intersection	O
,	O
and	O
GraphSLAM	O
.	O
</s>
<s>
SLAM	O
algorithms	O
are	O
based	O
on	O
concepts	O
in	O
computational	O
geometry	O
and	O
computer	B-Application
vision	I-Application
,	O
and	O
are	O
used	O
in	O
robot	B-General_Concept
navigation	I-General_Concept
,	O
robotic	O
mapping	O
and	O
odometry	O
for	O
virtual	B-Application
reality	I-Application
or	O
augmented	B-General_Concept
reality	I-General_Concept
.	O
</s>
<s>
Given	O
a	O
series	O
of	O
controls	O
and	O
sensor	O
observations	O
over	O
discrete	O
time	O
steps	O
,	O
the	O
SLAM	O
problem	O
is	O
to	O
compute	O
an	O
estimate	O
of	O
the	O
agent	B-General_Concept
's	O
state	O
and	O
a	O
map	O
of	O
the	O
environment	O
.	O
</s>
<s>
Like	O
many	O
inference	O
problems	O
,	O
the	O
solutions	O
to	O
inferring	O
the	O
two	O
variables	O
together	O
can	O
be	O
found	O
,	O
to	O
a	O
local	O
optimum	O
solution	O
,	O
by	O
alternating	O
updates	O
of	O
the	O
two	O
beliefs	O
in	O
a	O
form	O
of	O
an	O
expectation	B-Algorithm
–	I-Algorithm
maximization	I-Algorithm
algorithm	I-Algorithm
.	O
</s>
<s>
Statistical	O
techniques	O
used	O
to	O
approximate	O
the	O
above	O
equations	O
include	O
Kalman	O
filters	O
and	O
particle	B-Algorithm
filters	I-Algorithm
.	O
</s>
<s>
Bundle	B-General_Concept
adjustment	I-General_Concept
,	O
and	O
more	O
generally	O
maximum	B-General_Concept
a	I-General_Concept
posteriori	I-General_Concept
estimation	I-General_Concept
(	O
MAP	O
)	O
,	O
is	O
another	O
popular	O
technique	O
for	O
SLAM	O
using	O
image	O
data	O
,	O
which	O
jointly	O
estimates	O
poses	O
and	O
landmark	O
positions	O
,	O
increasing	O
map	O
fidelity	O
,	O
and	O
is	O
used	O
in	O
commercialized	O
SLAM	O
systems	O
such	O
as	O
Google	O
's	O
ARCore	O
which	O
replaces	O
their	O
prior	O
augmented	B-General_Concept
reality	I-General_Concept
computing	B-Device
platform	I-Device
named	O
Tango	B-Algorithm
,	O
formerly	O
Project	B-Algorithm
Tango	I-Algorithm
.	O
</s>
<s>
MAP	B-General_Concept
estimators	I-General_Concept
compute	O
the	O
most	O
likely	O
explanation	O
of	O
the	O
robot	O
poses	O
and	O
the	O
map	O
given	O
the	O
sensor	O
data	O
,	O
rather	O
than	O
trying	O
to	O
estimate	O
the	O
entire	O
posterior	O
probability	O
.	O
</s>
<s>
Topological	O
maps	O
are	O
a	O
method	O
of	O
environment	O
representation	O
which	O
capture	O
the	O
connectivity	O
(	O
i.e.	O
,	O
topology	B-Architecture
)	O
of	O
the	O
environment	O
rather	O
than	O
creating	O
a	O
geometrically	O
accurate	O
map	O
.	O
</s>
<s>
Location-tagged	O
visual	O
data	O
such	O
as	O
Google	O
's	O
StreetView	B-Application
may	O
also	O
be	O
used	O
as	O
part	O
of	O
maps	O
.	O
</s>
<s>
Essentially	O
such	O
systems	O
simplify	O
the	O
SLAM	O
problem	O
to	O
a	O
simpler	O
localization	B-General_Concept
only	O
task	O
,	O
perhaps	O
allowing	O
for	O
moving	O
objects	O
such	O
as	O
cars	O
and	O
people	O
only	O
to	O
be	O
updated	O
in	O
the	O
map	O
at	O
runtime	O
.	O
</s>
<s>
At	O
one	O
extreme	O
,	O
laser	O
scans	O
or	O
visual	O
features	O
provide	O
details	O
of	O
many	O
points	O
within	O
an	O
area	O
,	O
sometimes	O
rendering	O
SLAM	O
inference	O
unnecessary	O
because	O
shapes	O
in	O
these	O
point	O
clouds	O
can	O
be	O
easily	O
and	O
unambiguously	O
aligned	O
at	O
each	O
step	O
via	O
image	B-Algorithm
registration	I-Algorithm
.	O
</s>
<s>
At	O
the	O
opposite	O
extreme	O
,	O
tactile	B-Algorithm
sensors	I-Algorithm
are	O
extremely	O
sparse	O
as	O
they	O
contain	O
only	O
information	O
about	O
points	O
very	O
close	O
to	O
the	O
agent	B-General_Concept
,	O
so	O
they	O
require	O
strong	O
prior	O
models	O
to	O
compensate	O
in	O
purely	O
tactile	O
SLAM	O
.	O
</s>
<s>
Optical	O
sensors	O
may	O
be	O
one-dimensional	O
(	O
single	O
beam	O
)	O
or	O
2D	O
-	O
(	O
sweeping	O
)	O
laser	O
rangefinders	O
,	O
3D	O
high	O
definition	O
light	O
detection	O
and	O
ranging	O
(	O
lidar	O
)	O
,	O
3D	O
flash	O
lidar	O
,	O
2D	O
or	O
3D	O
sonar	B-Application
sensors	O
,	O
and	O
one	O
or	O
more	O
2D	O
cameras	O
.	O
</s>
<s>
Recent	O
approaches	O
apply	O
quasi-optical	O
wireless	O
ranging	O
for	O
multi-lateration	O
(	O
real-time	O
locating	O
system	O
(	O
RTLS	O
)	O
)	O
or	O
multi-angulation	B-Algorithm
in	O
conjunction	O
with	O
SLAM	O
as	O
a	O
tribute	O
to	O
erratic	O
wireless	O
measures	O
.	O
</s>
<s>
A	O
kind	O
of	O
SLAM	O
for	O
human	O
pedestrians	O
uses	O
a	O
shoe	O
mounted	O
inertial	O
measurement	O
unit	O
as	O
the	O
main	O
sensor	O
and	O
relies	O
on	O
the	O
fact	O
that	O
pedestrians	O
are	O
able	O
to	O
avoid	O
walls	O
to	O
automatically	O
build	O
floor	O
plans	O
of	O
buildings	O
by	O
an	O
indoor	B-Device
positioning	I-Device
system	I-Device
.	O
</s>
<s>
Early	O
implementations	O
of	O
this	O
technique	O
have	O
used	O
direction-of-arrival	O
(	O
DoA	O
)	O
estimates	O
of	O
the	O
sound	O
source	O
location	O
,	O
and	O
rely	O
on	O
principal	O
techniques	O
of	O
sound	O
localization	B-General_Concept
to	O
determine	O
source	O
locations	O
.	O
</s>
<s>
An	O
observer	O
,	O
or	O
robot	O
must	O
be	O
equipped	O
with	O
a	O
microphone	B-Algorithm
array	I-Algorithm
to	O
enable	O
use	O
of	O
Acoustic	O
SLAM	O
,	O
so	O
that	O
DoA	O
features	O
are	O
properly	O
estimated	O
.	O
</s>
<s>
drones	O
,	O
service	O
robots	O
)	O
,	O
it	O
is	O
valuable	O
to	O
use	O
low-power	O
,	O
lightweight	O
equipment	O
such	O
as	O
monocular	O
cameras	O
,	O
or	O
microelectronic	O
microphone	B-Algorithm
arrays	I-Algorithm
.	O
</s>
<s>
SLAM	O
with	O
DATMO	O
is	O
a	O
model	O
which	O
tracks	O
moving	O
objects	O
in	O
a	O
similar	O
way	O
to	O
the	O
agent	B-General_Concept
itself	O
.	O
</s>
<s>
For	O
example	O
,	O
this	O
can	O
be	O
done	O
by	O
storing	O
and	O
comparing	O
bag	B-General_Concept
of	I-General_Concept
words	I-General_Concept
vectors	O
of	O
scale-invariant	B-Algorithm
feature	I-Algorithm
transform	I-Algorithm
(	O
SIFT	O
)	O
features	O
from	O
each	O
previously	O
visited	O
location	O
.	O
</s>
<s>
Active	O
SLAM	O
is	O
generally	O
performed	O
by	O
approximating	O
the	O
entropy	B-Algorithm
of	O
the	O
map	O
under	O
hypothetical	O
actions	O
.	O
</s>
<s>
"	O
Multi	O
agent	B-General_Concept
SLAM	O
"	O
extends	O
this	O
problem	O
to	O
the	O
case	O
of	O
multiple	O
robots	O
coordinating	O
themselves	O
to	O
explore	O
optimally	O
.	O
</s>
<s>
Various	O
SLAM	O
algorithms	O
are	O
implemented	O
in	O
the	O
open-source	B-Application
software	I-Application
Robot	B-Application
Operating	I-Application
System	I-Application
(	O
ROS	B-Application
)	O
libraries	O
,	O
often	O
used	O
together	O
with	O
the	O
Point	B-Language
Cloud	I-Language
Library	I-Language
for	O
3D	O
maps	O
or	O
visual	O
features	O
from	O
OpenCV	B-Language
.	O
</s>
<s>
In	O
the	O
1990s	O
and	O
2000s	O
,	O
EKF	O
SLAM	O
had	O
been	O
the	O
de	O
facto	O
method	O
for	O
SLAM	O
,	O
until	O
the	O
introduction	O
of	O
FastSLAM	B-Application
.	O
</s>
<s>
The	O
acronym	O
SLAM	O
was	O
coined	O
within	O
the	O
paper	O
,	O
"	O
Localization	B-General_Concept
of	O
Autonomous	O
Guided	O
Vehicles	O
"	O
which	O
first	O
appeared	O
in	O
ISR	O
in	O
1995	O
.	O
</s>
<s>
Mass-market	O
SLAM	O
implementations	O
can	O
now	O
be	O
found	O
in	O
consumer	O
robot	O
vacuum	O
cleaners	O
and	O
virtual	B-Device
reality	I-Device
headsets	I-Device
such	O
as	O
the	O
Meta	B-General_Concept
Quest	I-General_Concept
2	I-General_Concept
and	O
PICO	B-Device
4	I-Device
for	O
markerless	O
inside-out	O
tracking	O
.	O
</s>
