Intern
ati
o
n
a
l Jo
urn
a
l
o
f
R
o
botics
a
nd Au
tom
a
tion
(I
JR
A)
Vol
.
3
,
No
. 2,
J
une
2
0
1
4
,
pp
. 75~
8
3
I
S
SN
: 208
9-4
8
5
6
75
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJRA
Actu
al F
light M
o
vem
e
nts
of El
ectric Heli
copters
for M
a
king a Dis
a
st
er Area M
o
nitorin
g
System
Ta
kuy
a
Sa
ito*
, Kenichi M
a
se**
* Res
ear
ch Ins
t
it
ute for
Natur
a
l
Hazards
&
Dis
a
s
t
er R
ecov
e
r
y
,
Ni
igat
a Univers
i
t
y
,
J
a
pan
** Ac
a
d
e
mic
Asse
mbly
, Ni
i
g
at
a
Uni
v
e
r
si
ty
, Ja
pan
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Sep 9, 2013
Rev
i
sed
Feb 2, 20
14
Accepted
Feb 26, 2014
This pap
e
r descr
i
bes the actual f
light
movements
of elec
tric
he
lic
opters an
d
how to implement a d
i
saster
area m
onitoring s
y
stem using electric ground
vehic
l
es
and
el
ectr
i
c h
e
li
copt
er
s
.
A quad-rotor
heli
copt
er and
a h
e
x-roto
r
helicopter wer
e
used in the proposed
sy
stem, specif
i
cally
,
the AR.Drone 2.0
and the DJI Innovations fr
ame kit with an
auto
pilot s
y
stem, respectiv
ely
.
W
e
develop
e
d a sof
t
ware fr
amework for
the AR.D
rone 2.0
,
which
makes it
possible to ob
ta
in fligh
t
da
ta
a
nd im
ages and
to contro
l au
to
m
a
tic fl
ight
operations b
y
co
mputer. The r
e
lationshi
p betwe
e
n
the flight p
a
ra
m
e
ters and
the real flight movements of
the
AR.Drone 2.0 were
experimentally
investigated
. In conclusion
, we found th
at the A
R
.Drone 2.0 with a software
framework and
a developed h
e
x
-rotor heli
copt
er
are s
u
ffi
cien
tl
y
effec
tive
in
implementing
a disaster
area
mon
itoring s
y
stem usi
ng el
ect
ric groun
d
vehicles and
electric helicopters
.
Keyword:
AR
.D
r
one
Disaster
Electric helicopter
Electric ve
hicle
Mo
n
itoring
syste
m
Copyright ©
201
4 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Taku
ya Saito
Research Institute for Nat
u
ral
Hazards a
n
d
Disaster Recovery, Niigata
Uni
v
ersity,
8
050
I
k
ar
ash
i
2
-
no
ch
o, N
i
shi-
ku
, N
iig
ata-
sh
i, N
iig
ata, 950
-21
8
1
,
Jap
a
n
Em
a
il: tak
u
y
a@tok
i
.wased
a.j
p
1.
INTRODUCTION
After a large-s
cale disaster,
s
u
ch as a m
a
jor
earthqua
k
e
, it is v
e
ry im
p
o
r
tan
t
to
assess th
e situ
atio
n
of
t
h
e di
sast
er si
t
e
as soon as p
o
ssi
bl
e. T
o
achi
e
ve t
h
i
s
, va
r
i
ous di
sast
e
r
area
m
oni
t
o
ri
ng
sy
st
em
s have
bee
n
pr
o
pose
d
.
F
o
r
exam
pl
e, som
e
use t
h
e se
nso
r
net
w
or
k t
e
c
h
n
i
que;
ho
we
ver
,
t
h
i
s
t
y
pe
of m
oni
t
o
ri
n
g
sy
st
e
m
can
onl
y
m
oni
t
o
r d
i
sast
er areas w
h
ere cam
eras were i
n
st
al
l
e
d i
n
ad
vance
.
Th
eref
ore
,
we
de
vel
o
ped a
n
d p
r
op
os
e
d
a di
sast
er area
m
oni
t
o
ri
ng sy
s
t
em
usi
ng
an el
ectric ground vehicle (EV) tha
t
can dri
v
e around the
disaste
r
site
t
o
su
r
v
ey
t
h
e s
i
t
u
at
i
on.
Thi
s
EV
has a
wi
rel
e
ss net
w
o
r
k
sy
st
em
and m
a
ny
sens
ors
,
i
n
cl
udi
ng
a
vi
de
o
cam
e
ra
and a
GPS syste
m
..
In
o
r
de
r t
o
co
m
m
uni
cat
e wi
t
h
eac
h
ot
he
r, t
h
e E
V
s a
r
e c
o
nnect
e
d
vi
a wi
rel
e
ss net
w
or
k
i
n
a
d
-
hoc
m
ode, such as
t
h
e el
ect
ri
c vehi
cul
a
r a
d
-
h
o
c
net
w
o
r
k (E
VA
NET
)
[
1
]
[
2
]
. Ho
weve
r, t
h
i
s
t
y
pe of m
oni
t
o
ri
n
g
syste
m
can onl
y survey areas
that are
visible
to the
dri
v
er,
and not places
hi
dden from
the dri
v
er, s
u
ch
as the
o
t
h
e
r sid
e
of a
h
i
gh
wall or the ro
of
o
f
an apartm
en
t b
u
ild
ing
.
Som
e
st
udi
es u
s
ed
unm
anne
d
aeri
a
l
vehi
cl
es
(U
AV
s)
e
qui
pped
with cam
eras to
s
u
rvey la
rge
disaster
areas
from
the sky. For e
x
a
m
ple, Ohm
i
nato
et al.
observed the
s
u
mmit areas
of active volcanoes
us
ing
a
UA
V [3]
.
Suz
uki
et al
.
propose
d
real-tim
e
hazard m
a
p gene
ra
tion using
a sm
all U
AV
[4]. Ale
x
is
et al
.
p
r
op
o
s
ed
aerial fo
rest fire su
rv
eillan
ce u
s
i
n
g UAVs
[5
].
Howev
e
r, certain
typ
e
s o
f
UAVs can
no
t b
e
op
erated
for a long tim
e
,
beca
use battery capacity is lim
ited. An
electric helicopte
r
(EHs
) ca
n
fly
fo
r ap
pr
o
x
im
ately
ten
minutes only.
For this
reas
on, alm
o
st
no E
H
can t
r
avel l
o
ng
distances.
There
f
ore,
we
propose a
disa
ster area m
onitori
ng sy
stem
using electric ground
ve
hicles
and electric
hel
i
c
opt
e
r
s
[6]
[
7]
. B
eca
use
E
V
s ca
n
r
u
n
l
o
ng
di
st
a
n
ces,
we ca
n
g
o
t
o
a di
sast
er
si
t
e
usi
n
g a
n
EV
a
n
d
t
h
e
n
launc
h the E
H
onsite. T
h
e E
V
serves as t
h
e aircraft ca
rri
er for t
h
e EH.
It charges t
h
e
battery of t
h
e
EH a
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3,
No
. 2,
J
u
ne 2
0
14:
7
5
– 8
3
76
t
r
ansm
i
t
s
t
h
e m
oni
t
o
ri
ng
dat
a
o
b
t
a
i
n
e
d
by
t
h
e E
H
t
o
ot
her
EV
s
or
t
o
base
st
at
i
ons
by
usi
n
g a
w
i
rel
e
ss
net
w
or
k, s
u
ch
as EVA
N
ET
. The EH ca
n ei
t
h
er fl
y
abo
v
e
t
h
e EV by
m
a
nual
o
p
e
r
at
i
on
or t
r
ac
k a m
ovi
ng E
V
autom
a
tically
by com
puter c
ont
rol.
A m
onitoring
syste
m
that com
b
ines electric grou
nd
vehi
cl
es and el
ect
ri
c hel
i
c
opt
ers
has seve
ral
adva
nt
age
s
o
v
e
r m
oni
t
o
ri
n
g
sy
st
em
s t
h
at
u
s
e onl
y
EVs
o
r
only EHs. Our
disaster
area m
o
n
ito
ring
syste
m
using electric ground ve
hicles
and electric
helicopte
r
s is at the early st
age of de
vel
opm
ent
.
I
n
t
h
i
s
pa
p
e
r, we
descri
be t
h
e d
e
vel
o
pm
ent
of t
h
e soft
ware f
r
am
ework
fo
r
t
h
e AR
.
D
r
o
n
e
2.
0, t
h
e fl
i
ght
-
d
at
a o
b
t
a
i
n
ed
by
o
u
r
pr
ot
ot
y
p
e
,
a
n
d
t
h
e
fl
i
g
ht
per
f
o
r
m
a
nce
of
t
h
e hex
-
r
o
t
o
r hel
i
c
opt
e
r
t
h
at
we d
e
vel
o
ped
.
Thi
s
pape
r i
s
or
ga
ni
zed a
s
fol
l
o
ws:
In
Sect
i
on
II
, w
e
desc
ri
be t
h
e
speci
fi
cat
i
o
ns
of
el
ect
ri
c
hel
i
c
opt
e
r
s re
qui
red
by
t
h
e di
sast
er area
m
oni
t
o
ri
n
g
s
y
st
em
. In Sect
i
on II
I,
we di
scuss t
h
e soft
war
e
fram
e
wor
k
f
o
r
t
h
e AR
.D
r
one
2.0 a
n
d h
o
w i
t
can be use
d
t
o
de
vel
o
p a m
oni
t
o
ri
n
g
sy
st
em
[8]
.
In Sect
i
on
IV
,
we desc
ri
be
ho
w t
o
co
nt
r
o
l
t
h
e fl
i
ght
o
f
t
h
e AR
.D
r
one
2.
0.
In Sect
i
o
n V,
we sh
o
w
t
h
e expe
ri
m
e
nt
al
resul
t
s
,
i
n
cl
udi
ng
t
h
e
f
l
i
ght
dat
a
o
f
t
h
e AR
.
D
r
o
ne
2.
0 a
n
d
t
h
e
fl
i
g
h
t
per
f
o
r
m
a
nce of t
h
e c
u
st
om
hex
-
r
o
t
o
r
hel
i
c
opt
e
r
.
Sect
i
on
V
I
c
o
n
c
l
udes
t
h
e
pa
p
e
r.
2.
REQUIRE
M
ENTS AND SPECIFICATIO
NS OF ELECTRIC
HELICOPTE
R
S
We a
d
opted electric helicopters in
our
disa
ster ar
ea m
oni
t
o
ri
ng
sy
st
em
and
deci
ded t
o
use a
n
E
V
t
o
p
r
ov
id
e th
e energ
y
n
e
ed
ed
by th
e h
e
lico
p
t
ers. Fo
r th
is
s
t
udy, the electric helicopt
er
m
u
st satisfy various
requ
irem
en
ts as fo
llo
ws:
-
Th
e h
e
licop
ter m
u
st
b
e
easy
to
o
p
e
rate witho
u
t
sp
ecial
trai
n
i
ng
.
-
The
hel
i
c
o
p
t
e
r
m
u
st
be co
nt
ro
l
l
a
bl
e by
c
o
m
put
er
vi
a a
wi
rel
e
ss net
w
o
r
k.
-
Th
e
h
e
licop
ter
m
u
st b
e
ab
le to send
real-
tim
e
cam
era im
age
s
via a
wireless
network.
-
Using GPS or im
age analysis, the helicopt
e
r m
u
st be
able to fly above
an el
ectric ground ve
hicle and
track
it au
to
m
a
tically.
A h
e
licop
ter th
at satisfies al
l o
f
th
ese d
e
man
d
s is n
o
t
cu
rren
tly av
ailab
l
e co
mmerciall
y. However,
we
fo
u
n
d
t
h
e
AR
.D
r
one
2
.
0
hel
i
c
opt
e
r
,
w
h
i
c
h sat
i
s
fi
es
m
o
st
of
t
h
ese
de
m
a
nds.
T
h
ere
f
ore
,
we a
d
opt
e
d
t
h
e
AR
.D
r
one
2.
0
as o
n
e o
f
t
h
e
EHs
fo
r t
h
e
di
sast
er area m
oni
t
o
ri
ng sy
st
e
m
. An AR
.
D
r
o
ne 2
.
0 i
s
sh
ow
n o
n
t
h
e
l
e
ft
si
de
of
Fi
g
u
re
1
.
Fi
gu
re 1.
AR
.
D
r
o
n
e
2.
0
a
n
d
t
h
e
he
x
-
r
o
t
o
r h
e
l
i
c
opt
er
The
AR
.
D
r
o
ne
2.
0
has
m
a
ny
excel
l
e
nt
f
eat
u
r
es, i
n
cl
u
d
i
n
g
a Li
n
u
x
-
based
em
bedde
d
sy
st
em
. It
has a
n
IEEE
802.11
b/g wireless
net
w
ork card,
which ca
n c
o
nn
ect
to a
nd c
o
m
m
u
n
icate with
personal com
pute
r
s
via
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
Real
Fl
i
g
ht
M
o
vement
s
of
El
e
c
t
r
i
c
Hel
i
c
opt
e
rs f
o
r
Maki
ng
a
Di
sast
er
Are
a M
o
ni
t
o
ri
n
g
…
(
T
akuy
a
S
a
i
t
o
)
77
an ad
-h
oc net
w
o
r
k
.
It
has
several
u
s
ef
ul
senso
r
s, i
n
cl
udi
ng a hi
gh
-
d
efi
n
i
t
i
on vi
de
o cam
era, t
h
ree-axi
s
gyroscopes
, three-axis accele
r
om
eters,
three-axis m
a
gnetometers, a
pre
ssure
sens
or, a
n
d a
n
ultrasound se
nsor
for g
r
o
und
altitu
d
e
m
easu
r
emen
t.
Th
e AR.Dron
e
2
.
0
h
a
s
a
p
u
b
licly av
ailab
l
e set
o
f
APIs th
at can b
e
used
to
p
r
og
ramm
a
tica
lly o
p
e
rate th
e h
e
lico
p
t
er v
i
a wireless n
e
t
w
o
r
k;
t
h
ere
f
ore,
we de
vel
o
ped
com
put
er so
ft
ware t
o
do
so
.
Because the
AR.Drone
2.0 has th
e
de
fe
cts
m
e
ntioned above, we
decided t
o
de
velop a
n
othe
r
hel
i
c
opt
e
r
– a
hex
-
r
o
t
o
r
hel
i
c
opt
e
r
usi
n
g t
h
e
DJ
I I
n
no
vat
i
o
n F
5
50
f
r
am
e
ki
t
an
d t
h
e
Wo
oK
o
n
g
-
M
a
u
t
o
pi
l
o
t
sy
st
em
[9]
.
Th
e he
x-
rot
o
r
hel
i
copt
er
i
s
s
h
o
w
n
on
t
h
e
ri
g
h
t
s
i
de o
f
Fi
gu
re
1
.
Thi
s
he
x-
r
o
t
o
r hel
i
c
opt
er
sat
i
sfi
e
s
all o
f
ou
r
req
u
i
rem
e
n
t
s, an
d it can
b
e
op
erated
v
e
ry easily usin
g its au
to p
i
l
o
t system
.
3.
DEVELOP
M
ENT OF
THE
MO
NITO
RI
NG
SY
STEM
USI
N
G
A
R
.
D
RO
NE 2
.
0
Parr
ot
,
Inc
.
de
vel
o
ped
an
S
DK
f
o
r t
h
e A
R
.Dr
o
ne, a
n
d
t
h
e so
ur
ce co
de i
s
o
p
e
n
t
o
t
h
e p
u
b
l
i
c
.
Howev
e
r, th
is
SDK con
s
ists
o
f
a larg
e nu
m
b
er
o
f
file
s, and
its stru
cture
is to
o
co
m
p
lex to
un
d
e
rstand with
i
n
su
ffi
ci
ent
do
cum
e
nt
at
i
on. T
h
ere
f
o
r
e, t
h
e s
o
u
r
ce c
ode i
s
t
oo
di
f
f
i
c
ul
t
t
o
m
odi
fy
t
o
im
plem
ent
t
h
e fu
nc
t
i
ons
t
h
at
we
nee
d
.
I
n
a
d
di
t
i
on,
t
h
i
s
SD
K
was
pri
m
ari
l
y
desi
gne
d t
o
t
e
st
fu
nct
i
ons
du
ri
n
g
t
h
e
de
vel
o
pm
ent
of
t
h
e
AR
.D
r
one;
t
h
e
r
ef
ore
,
i
t
i
s
not
sui
t
a
bl
e fo
r o
u
r m
oni
t
o
ri
ng
sy
st
em
. Inst
ead, we
deci
de
d
t
o
de
vel
o
p fra
m
e
wo
r
k
soft
ware
t
o
c
o
nt
r
o
l
t
h
e
AR
.
D
ro
ne
2.
0.
The AR
.Drone
2.0 ca
n c
o
nne
c
t to pers
onal
com
put
ers via
its IEEE 802.11 b/
g wireless
network in
ad-
h
oc m
ode. The IP a
d
dress
of t
h
e AR
.D
r
one
2.
0 i
s
19
2.
16
8.
1.
1. T
h
e I
P
add
r
ess
of t
h
e pe
rso
n
al
co
m
put
er
t
h
at
con
n
ect
s
t
o
t
h
e AR
.D
r
one
2.
0 i
s
ass
i
gne
d aut
o
m
a
ti
cal
l
y
by
t
h
e DHC
P
ser
v
er
fr
om
192.
1
6
8
.
1.
2 t
o
19
2.
1
6
8
.
1
.
5
.
The AR
.
D
r
o
ne 2.
0 uses se
ver
a
l
port
s
t
o
co
m
m
uni
cat
e wi
th t
h
e com
put
er
s. The com
m
uni
cat
i
o
n
po
rt
s a
r
e s
h
o
w
n i
n
Ta
bl
e 1
.
The
AR.Drone
2.0 uses UDP port
5556 to r
eceive
flight
ope
ration co
mmands. Every 30 m
s
, the
pers
o
n
al
com
put
er se
nds
AT
com
m
a
nds as
ASC
I
I st
ri
n
g
s
t
o
t
h
e AR
.
D
ro
ne 2
.
0 t
o
i
ndi
c
a
t
e
fl
i
ght
m
ovem
e
nt
s,
suc
h
as t
a
keo
f
f
,
l
a
ndi
n
g
, asce
nd
, desce
n
d, pi
t
c
h, r
o
l
l
,
and y
a
w, an
d t
o
co
n
f
i
g
ure t
h
e set
t
i
ngs
of t
h
e
AR
.
D
r
o
n
e
2
.
0
.
Nav
d
a
ta, su
ch
as v
e
lo
cit
y
,
altitu
d
e
, p
itch
,
ro
ll,
and
y
a
w, are
o
b
t
ained
b
y
th
e
AR.Dro
ne 2.0 th
rou
g
h
its
sens
ors
an
d se
nt
t
o
t
h
e pe
rs
o
n
al
com
put
er t
h
r
o
ug
h
UD
P
p
o
rt
55
5
4
a
b
o
u
t
1
5
t
i
m
es per secon
d
.
Th
e AR
.Dr
o
ne
2.
0 al
so se
n
d
s
vi
de
o dat
a
t
h
r
o
ug
h TC
P
p
o
rt
55
5
5
. T
h
i
s
vi
d
e
o dat
a
i
s
enc
o
ded i
n
t
h
e
H.
2
64
or M
P
EG
4
fo
rm
at
with
Parro
t
’s
orig
in
al
h
e
ad
er
called
PaVE. Fin
a
lly, TCP
p
o
r
t
555
9 is
u
s
ed to
send
an
y co
nf
igu
r
ation data f
o
r
t
h
e AR
.D
ro
ne
2.
0.
Thi
s
s
o
ft
ware
fram
e
wor
k
fo
r
t
h
e fl
i
ght
c
o
n
t
rol
of
t
h
e AR
.Dr
o
ne 2.
0 wa
s
de
vel
o
pe
d o
n
M
a
c
O
S X
10
.8
.2 M
o
u
n
t
a
i
n
Li
o
n
. T
h
e
st
ruct
u
r
e
of t
h
i
s
so
ft
wa
re f
r
am
ework i
s
sho
w
n i
n
Fi
g
u
re
2. T
h
e s
o
ft
ware
fram
e
work is c
o
m
posed of
five threa
d
s, bec
a
use t
h
e data
at each port need continuous processing.
Tabl
e
1. C
o
m
m
uni
cat
i
on P
o
r
t
s of
AR
.
D
ro
ne
2.
0
Por
t
Dir
ection
Pr
otocol
Descr
iption
5554
Drone to PC
UDP
Flight data (navdata)
5555
Dr
one to PC
T
C
P
Video data
5556
PC to Dr
one
UDP
Oper
ate co
m
m
and
(
A
T)
5559
Dr
one to PC
T
C
P
Contr
o
l por
t
Fi
gu
re 2.
St
r
u
c
t
ure of
t
h
e so
ft
ware
f
r
am
ewor
k
Main Program
AT
C
o
mmand
Sender
UDP 5556
Navdata
Receive
r
UDP 5554
Video
Receive
r
and
Decoder
TCP 5555
Config
Data
Receive
r
TCP 5559
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3,
No
. 2,
J
u
ne 2
0
14:
7
5
– 8
3
78
4.
HOW
T
O
CO
NTROL
THE
FLIGHT OF AR
.D
RO
NE 2
.
0
4.
1.
Fl
i
g
h
t
Da
ta
o
f
AR
.Dr
o
n
e
2.
0
The s
o
ft
ware
fram
ework that
we
devel
ope
d ena
b
les us
t
o
receive
navdat
a, including the flight
data,
from
the AR.Drone 2.0.
An exam
pl
e of actual flight da
ta received fro
m
the AR.Drone
2.0 is shown in
Fi
gu
re 3.
We set
t
h
e AR
.Dr
o
ne 2.
0 t
o
na
vdat
a
_de
m
o
m
ode;
t
h
erefo
r
e, i
t
sends
nav
d
at
a abo
u
t
15 t
i
m
e
s per
secon
d
. Nav
d
a
ta in
clu
d
e
s
b
a
ttery lev
e
l, p
itch, yaw, altit
u
d
e
, th
ree-ax
is v
e
l
o
city, an
d
o
t
h
e
r in
fo
rm
atio
n
.
Fig
u
re
4 s
h
o
w
s
t
h
e a
s
si
gnm
ent
of
t
h
e x,
y
,
z,
pi
t
c
h,
rol
l
,
a
n
d y
a
w
a
x
es
of
t
h
e
AR
.
D
r
o
n
e
2.
0.
4.2.
Flight P
a
ramet
er
s of AR.Drone 2.0
The fl
i
g
ht
of
t
h
e AR
.D
r
one
2.0 i
s
cont
rol
l
ed by
AT co
m
m
a
nds, whi
c
h are ASC
II s
t
ri
ngs
. The
AT
*R
EF c
o
m
m
a
nd i
s
u
s
ed t
o
co
nt
r
o
l
t
h
e
b
a
si
c beha
vi
o
r
,
suc
h
as t
a
ke
of
f
,
l
a
ndi
n
g
, em
ergency
st
op
, an
d reset
.
The
AT*
P
C
M
D com
m
and cont
rol
s
t
h
e
fl
i
ght
m
o
t
i
ons, s
u
ch a
s
pi
t
c
h
,
rol
l
,
a
nd y
a
w.
The
val
u
e
of
eac
h
param
e
t
e
r i
s
a pe
rcent
a
ge
of
t
h
e m
a
xim
u
m
be
ndi
ng
an
gl
e
,
w
h
i
c
h
i
s
set
t
o
an
ot
he
r c
o
n
f
i
g
urat
i
o
n
para
m
e
t
e
r
called the E
u
le
r a
ngle m
a
x (E
AM). There
f
ore, a targ
et-be
n
ding a
ngle c
a
n
be calculate
d
by the following:
Tar
g
et
A
ngl
e
(r
ad)
EAM
(r
ad
) ×
Pitch
(%
)
(1
)
For exam
ple, if we set the value of t
h
e EAM para
m
e
t
e
r to 0
.
5 a
nd t
h
e v
a
l
u
e of t
h
e pi
t
c
h pa
ram
e
t
e
r
t
o
0
.
3
,
t
h
e
t
a
r
g
et
-be
ndi
ng
pi
t
c
h a
ngl
e i
s
0
.
5
r
a
d ×
3
0
%
0
.
15
r
a
d.
The de
vel
o
pe
r
m
a
nual
of t
h
e AR
.D
ro
ne
2
.
0 [
1
0]
expl
ai
ns t
h
ese
para
m
e
t
e
rs as abo
v
e. H
o
wev
e
r,
whe
n
we set these two pa
ra
meters to a specific target
value, the drone did not fly at that target angle
.
There
f
ore,
we
per
f
o
r
m
e
d a fl
i
ght
ex
peri
m
e
nt
wi
t
h
t
h
e
AR
.
D
r
o
n
e
2.
0
t
o
i
nve
st
i
g
at
e
the actual effect of these
param
e
ters.
Fi
gu
re 3.
Act
u
al
fl
i
ght
dat
a
re
cei
ved fr
om
t
h
e
AR
.
D
r
o
ne 2.
0
Fi
gu
re 4.
X
,
Y,
Z
a
x
i
s
an
d pi
t
c
h, r
o
l
l
,
y
a
w
a
x
i
s
di
rect
i
o
n of
t
h
e AR
.
D
r
one
2.
0
Navdata Received! packet
No=360 size=500 header=143277
863
2
state=260048
08
0 seqNu
m
=368370 vi
sionFlag=0
id=0 size=148 -
N
avdata_dem
o
-
ctr
l
_
s
tate=131072 vBat%=83
phi=-
133 psi=-
222
67 alt
itude=0 vx=0 vy
=0 vz=0
id=16 size=328 -
V
ision Detect-
id=6553
5 size=8 -
C
heck Su
m
-
check
su
m
=
7432
X
Y
Z
Y
Y
a
a
w
w
R
R
o
o
l
l
l
l
P
P
i
i
t
t
c
c
h
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
Real
Fl
i
g
ht
M
o
vement
s
of
El
e
c
t
r
i
c
Hel
i
c
opt
e
rs f
o
r
Maki
ng
a
Di
sast
er
Are
a
M
o
ni
t
o
ri
n
g
…
(
T
akuy
a
S
a
i
t
o
)
79
5.
EX
P
E
R
I
M
E
NT
5.1.
Experimental Envir
o
nment
Ou
r fl
i
g
ht
ex
p
e
ri
m
e
nt
i
nvest
i
g
at
ed t
h
e rel
a
t
i
ons
hi
p bet
w
e
e
n t
h
e pa
ram
e
t
e
r val
u
es a
n
d
real
fl
i
ght
m
ovem
e
nt
s of
t
h
e AR
.D
ro
ne
2.
0.
The
ex
peri
m
e
nt
al
envi
r
o
n
m
ent
i
s
sh
ow
n
i
n
t
a
bl
e
2.
Tabl
e
2. E
x
per
i
m
e
nt
al
Envi
r
o
nm
ent
OS
M
acOS X 10.
8 M
o
untain L
i
on
Co
m
puter
Apple M
acBook Air
13-
inch M
i
d 2012
Pro
cesso
r
In
tel Co
re i7
M
e
m
o
ry
8GB DDR3L
SDR
A
M
Stor
age SSD
256GB
RC Helicopter
AR.
D
r
one 2.
0
AR.
D
r
one Har
d
war
e
Ver
s
ion 2.
1
AR.
D
r
one Software
Ver
s
ion 2.
2.
9
5.
2.
Fl
i
g
h
t
E
x
peri
ment of AR.Drone 2.0
We c
h
o
s
e t
h
e
t
a
rget
-
b
e
n
di
n
g
pi
t
c
h a
n
gl
e
t
o
be
0
.
2
ra
d.
The
t
a
r
g
et
-be
ndi
ng
pi
t
c
h a
ngl
e ca
n
b
e
calculated by
form
ula (1).
We cha
nge
d
t
h
e
co
m
b
in
atio
n
o
f
th
e EAM v
a
l
u
e and
th
e
p
itch v
a
lu
e
fo
ur times, bu
t
al
l
of t
h
e c
o
m
b
i
n
at
i
on
val
u
es
resul
t
e
d i
n
t
h
e
sam
e
t
a
rget
-be
ndi
ng a
n
gl
e, t
h
at
i
s
, 0.
2 ra
d.
The c
o
m
b
i
n
at
ions
o
f
t
h
e val
u
es
of
t
h
e EAM
a
n
d t
h
e
pi
t
c
h
pa
ram
e
ters a
r
e s
h
o
w
n i
n
Ta
bl
e
3.
The fl
i
g
ht
dat
a
of t
h
e AR
.
D
r
o
ne 2.
0 can be
obt
ai
ne
d by
re
fer
r
i
n
g t
o
Nav
d
at
a, w
h
i
c
h wa
s sent
every
65
m
s
. The e
x
peri
m
e
nt
was
per
f
o
r
m
e
d fi
ve
t
i
m
e
s, and
t
h
e
fi
nal
res
u
l
t
wa
s o
b
t
a
i
n
e
d
by
a
v
era
g
i
n
g t
h
em
. I
n
t
h
e
expe
ri
m
e
nt
, w
e
l
a
unc
he
d t
h
e
AR
.D
r
one
2
.
0
and
,
few
seco
n
d
s l
a
t
e
r,
we
set
t
h
e E
A
M
a
n
d
t
h
e pi
t
c
h
v
a
l
u
e
s
.
Fi
gu
re
5 s
h
ow
s t
h
e
rel
a
t
i
ons
h
i
p bet
w
een
t
h
e
param
e
t
e
r val
u
es a
n
d
t
h
e act
ual
fl
i
g
ht
pi
t
c
h
an
gl
es se
nt
fro
m
th
e AR.Dro
ne 2
.
0
.
We set th
e EAM an
d
th
e
p
itch
v
a
lu
es at ti
m
e
= 0
.
All co
m
b
in
atio
ns o
f
EAM an
d
pi
t
c
h a
ngl
es
w
e
re ex
pect
ed
t
o
res
u
l
t
i
n
t
h
e
sam
e
t
a
rget
angl
e,
w
h
i
c
h i
s
0.
2 ra
d.
H
o
w
e
ver
,
t
h
e e
x
p
e
ri
m
e
nt
sho
w
e
d
t
h
at
t
h
e AR
.
D
r
one
2
.
0 fl
i
e
s at
di
ffe
rent
a
ngl
es
, w
h
i
c
h a
r
e i
n
pr
o
p
o
r
t
i
o
n t
o
t
h
e
EAM
val
u
es.
Whe
n
EAM
i
s
0.2
,
t
h
e fl
i
ght
pi
t
c
h angl
e i
s
aro
u
nd
0.
2 rad
.
Li
ke
wi
se, whe
n
EAM
i
s
0.3, 0
.
4
,
an
d 0.
5, t
h
e fl
i
g
h
t
pi
t
c
h
an
g
l
e is aro
und 0
.
3
,
0.4, an
d 0.5
r
a
d
,
r
e
sp
ectiv
ely.
Tabl
e
3. C
o
m
b
i
n
at
i
ons
o
f
E
u
l
e
r
An
gl
e M
a
x
and
Pi
t
c
h
Para
m
e
t
e
rs
E
u
ler
Angle
M
a
x (r
ad)
Pitch (
%
)
T
a
r
g
et
Angle (
r
a
d)
0.
2 1.
0
0.
2
0.
3 0.
65
0.
195
0.
4 0.
5
0.
2
0.
5 0.
4
0.
2
Fi
gu
re 6 s
h
o
w
s t
h
e rel
a
t
i
ons
hi
p bet
w
een t
h
e param
e
t
e
r val
u
es an
d t
h
e act
ual
fl
i
ght
sp
eeds.
Whe
n
EAM
i
s
0.
2
0
, t
h
e act
ual
fl
i
g
ht
spee
d i
n
crea
se
s gra
d
ual
l
y
. H
o
we
ve
r,
whe
n
EAM
i
s
0.
3
0
, t
h
e act
ual
fl
i
g
ht
spee
d
increases
som
e
what
ra
pidly a
n
d ove
r
s
hoo
ts slig
h
tly. Wh
en
EAM
is 0
.
40
an
d 0.50, the
flight s
p
eed inc
r
eases
very
rapi
dl
y
an
d ove
rs
ho
ot
s g
r
eat
l
y
.
The res
u
lts
are as
follows:
-
The
flight
angl
e does
not m
a
tch the
Target
Angle calcu
lat
e
d
b
y
(1
), ex
cep
t
th
at the EAM v
a
lu
e is 0.2.
-
The flight angl
es
agree
s
o
m
e
what with
t
h
e EAM values
.
-
W
h
en
t
h
e EAM v
a
lu
e is 0.2
an
d 0.3, th
e act
u
a
l fligh
t
p
itch
an
g
l
e is stab
le.
-
Wh
en
t
h
e EAM v
a
lu
e is 0.4
an
d 0.5, th
e act
u
a
l fligh
t
p
itch
an
g
l
e is un
stable.
-
Wh
en
t
h
e EAM v
a
lu
e is 0.2, th
e
fl
i
ght
s
p
ee
d i
n
c
r
eases
g
r
a
dual
l
y
.
-
Whe
n
t
h
e E
A
M
val
u
e
i
s
0
.
3
or
hi
ghe
r,
fl
i
g
h
t
spee
d
ove
rs
h
oot
i
s
o
b
se
rve
d
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3,
No
. 2,
J
u
ne 2
0
14:
7
5
– 8
3
80
Fi
gu
re
5.
R
e
l
a
t
i
ons
hi
p
bet
w
ee
n
param
e
t
e
r va
l
u
es a
n
d
fl
i
g
ht
pi
t
c
h a
ngl
es
Fi
gu
re
6.
R
e
l
a
t
i
ons
hi
p
bet
w
ee
n
param
e
t
e
r va
l
u
es a
n
d
fl
i
g
ht
spee
ds
In c
o
ncl
u
si
on
,
we f
o
un
d t
h
at
t
h
e act
ual
fl
i
g
ht
pi
t
c
h a
n
gl
e ag
rees s
o
m
e
what
wi
t
h
t
h
e E
A
M
val
u
e
,
a
n
d
the ideal EAM value is 0.2, because, wh
e
n
EAM is 0.3
or
higher, the flight
angle and the flight spee
d becom
e
unst
a
bl
e.
We a
l
so f
o
un
d t
h
at
t
h
e
pi
t
c
h
val
u
e
has
n
o
ef
fect
o
n
t
h
e
fl
i
g
ht
a
n
g
l
es w
h
en
t
h
e
v
a
l
u
e i
s
0.
4
or
h
i
ghe
r.
5.
3.
Fl
i
g
h
t
E
x
peri
ment w
i
t
h
E
A
M V
a
l
u
e
Fi
xed
at
0
.
2
We deci
de
d t
o
set
t
h
e EAM
val
u
e at
0.2, b
e
c
a
use we f
o
u
n
d
t
h
at
0.2 i
s
t
h
e i
d
eal
EAM
val
u
e based o
n
t
h
e ab
o
v
e-m
e
nt
i
oned
ex
peri
m
e
nt
.
Ho
we
ver,
t
h
e res
u
l
t
s
al
so s
h
o
w
t
h
at
t
h
e pi
t
c
h
val
u
e
has
no
ef
fect
on t
h
e
actual flight a
ngle.
We s
u
s
p
ected that
a pi
t
c
h val
u
e
of
1.
0 d
o
es
n
o
t
m
e
an 1
0
0
%, a
n
d
t
h
at
a val
u
e
o
f
0.
4 o
r
higher m
eans ove
r
100%.
T
h
ere
f
ore, to determine wh
at
t
h
e
pi
t
c
h
val
u
e m
eans,
we
pe
rf
orm
e
d a
fl
i
ght
expe
ri
m
e
nt
wh
ere t
h
e
EAM
v
a
l
u
e wa
s fi
xed
at
0.
2 a
n
d
t
h
e
p
i
t
c
h val
u
e c
h
an
ged
t
o
0.
1,
0
.
2
,
0.
3,
0
.
4
,
a
n
d
1
.
0.
0
0.
1
0.
2
0.
3
0.
4
0.
5
0.
6
0.
7
0.
8
0
0
.
6
5
1
.
3
1.
95
2.
6
3
.
2
5
3
.
9
4.
55
5.
2
Flig
ht
Pit
c
h
Angle
(r
a
d
)
Time
(s
ec)
EAM=0.
20
/
Pitc
h=1.
0
EAM=0.
30
/
Pitc
h=0.
65
EAM=0.
40
/
Pitc
h=0.
50
EAM=0.
50
/
Pitc
h=0.
40
‐
10
‐
5
0
5
10
15
20
0
0
.
6
5
1
.
3
1.
95
2.
6
3
.
2
5
3
.
9
4.
55
5.
2
Fligh
t
Speed
(km/h)
Time
(sec
)
EAM=0.
20
/
Pitc
h=1.
0
EAM=0.
30
/
Pitc
h=0.
65
EAM=0.
40
/
Pitc
h=0.
50
EAM=0.
50
/
Pitc
h=0.
40
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
Real
Fl
i
g
ht
M
o
vement
s
of
El
e
c
t
r
i
c
Hel
i
c
opt
e
rs f
o
r
Maki
ng
a
Di
sast
er
Are
a
M
o
ni
t
o
ri
n
g
…
(
T
akuy
a
S
a
i
t
o
)
81
Fi
gu
re 7.
R
e
l
a
t
i
ons
hi
p
bet
w
ee
n pi
t
c
h val
u
e
a
n
d
fl
i
g
ht
pi
t
c
h angl
e
Fi
gu
re
8.
R
e
l
a
t
i
ons
hi
p
bet
w
ee
n
pi
t
c
h
val
u
e
a
n
d
fl
i
g
ht
s
p
eed
Fi
gu
re
7 s
h
ow
s t
h
e
rel
a
t
i
ons
h
i
p bet
w
een
pi
t
c
h
val
u
es
an
d
fl
i
ght
pi
t
c
h a
n
gl
es.
Wh
en
we
cha
nge
d t
h
e
pi
t
c
h
val
u
e t
o
0.
1,
0.
2, a
n
d
0
.
3, t
h
e
fl
i
ght
p
i
t
c
h an
gl
es ch
ange
d l
i
n
ea
rl
y
.
Ho
we
ver
,
w
h
en w
e
set
t
h
e
pi
t
c
h
val
u
e t
o
0
.
4
or
hi
g
h
er
, t
h
e ac
t
u
al
fl
i
ght
pi
t
c
h an
gl
e
d
i
d
n
o
t ch
ang
e
lin
early; in
stead
, it see
m
ed to exce
ed t
h
e
max
i
m
u
m
lev
e
l. Th
is
resu
lt sh
ows t
h
at th
e
v
a
lu
e
o
f
th
e
p
i
tch
p
a
ram
e
ter i
s
v
a
lid
wh
en
it
is less th
an or equ
a
l
t
o
0.
3.
We al
so o
b
ser
v
e
d
t
h
a
t
whe
n
t
h
e pi
t
c
h val
u
e i
s
0.
1,
t
h
e fl
i
ght
pi
t
c
h
angl
e i
s
ar
ou
n
d
0.
0
5
ra
d,
wh
i
c
h i
s
app
r
oxi
m
a
t
e
ly
25
% of t
h
e EA
M
val
u
e o
f
0.
2
rad. Li
kewi
se,
whe
n
t
h
e pi
t
c
h val
u
e
s
are 0
.
2 an
d 0.
3, t
h
e fl
i
ght
pi
t
c
h a
ngl
es a
r
e ap
pr
oxi
m
a
t
e
ly
0.1
0
rad
an
d
0.
15
ra
d,
res
p
ect
i
v
el
y
,
whi
c
h are a
p
pr
o
x
i
m
at
ely
50% a
n
d 7
5
%
,
respectively, of the
EAM
val
u
e.
Fi
gu
re 8 s
h
ow
s t
h
e rel
a
t
i
ons
hi
p
bet
w
ee
n p
i
t
c
h val
u
es a
n
d fl
i
g
ht
spee
ds
. The fl
i
g
ht
s
p
eed can
be
measured whe
n
the
pitch
va
lue is gr
eat
er t
h
an or
e
q
ual
t
o
0.
3.
Al
t
h
o
u
gh t
h
e m
easured flight s
p
ee
ds are
aro
u
nd
0
wh
en
t
h
e pi
t
c
h
val
u
es are
0.
1 a
nd
0.
2, t
h
e AR
.D
r
one
2
.
0 act
ual
l
y
fl
i
e
s fo
rwa
r
d
.
Thi
s
m
eans t
h
at
t
h
e
flig
h
t
sp
eed
d
a
ta in
th
e
n
a
vd
ata is un
reliab
l
e,
esp
ecially
wh
en
th
e p
itch
v
a
lu
e is less th
an
o
r
equ
a
l to 0.2.
In con
c
lu
si
o
n
,
th
e exp
e
rim
e
n
t
al resu
lt sho
w
s th
e
fo
llowing
:
-
The E
A
M
val
u
e i
s
t
h
e m
a
xim
u
m
bendi
ng
pi
t
c
h a
ngl
e.
-
The
pi
t
c
h
val
u
e i
s
t
h
e
perc
ent
a
ge
of
t
h
e m
a
xim
u
m
bendi
n
g
pi
t
c
h a
ngl
e
.
‐
0.
05
0
0.
05
0.
1
0.
15
0.
2
0.
25
0
1
.3
2
.
6
3
.9
5
.
2
6
.5
7
.
8
9
.1
Fligh
t
Pitch
An
gle
(ra
d
)
Time
(sec
)
Pitc
h=0.
1
Pitc
h=0.
2
Pitc
h=0.
3
Pitc
h=0.
4
Pitc
h=1.
0
‐
2
0
2
4
6
8
10
12
0
1
.3
2
.
6
3
.9
5
.
2
6
.5
7
.
8
9
.1
Fligh
t
Speed
(km/h)
Time
(sec
)
Pitc
h=0.
1
Pitc
h=0.
2
Pitc
h=0.
3
Pitc
h=0.
4
Pitc
h=1.
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
089
-48
56
IJR
A
V
o
l
.
3,
No
. 2,
J
u
ne 2
0
14:
7
5
– 8
3
82
-
A
p
itch
v
a
lu
e of
0
.
1
co
rr
espond
s t
o
ab
ou
t
2
5
% of
th
e EA
M
v
a
lu
e.
-
Val
i
d
pi
t
c
h
val
u
es a
r
e
fr
om
0 t
o
0
.
3
.
These e
x
perimental res
u
lts re
vealed t
h
at the
actual
t
a
rget
-
b
endi
ng
fl
i
g
ht
p
i
t
c
h an
gl
e of
t
h
e AR
.
D
r
one
can
b
e
calcu
lated
b
y
th
e fo
llowing
:
Tar
g
et
A
ngl
e (rad)
EAM
(r
ad
) ×
25
×
Pitch
(%
)
(2
)
5.4. Flight
Per
f
orm
a
nce
Me
asureme
nt
of the Hex-rotor Helicopter
We m
easured
t
h
e fl
i
g
ht
pe
rf
orm
a
nce o
f
t
h
e he
x-
r
o
t
o
r
hel
i
c
opt
e
r
t
h
at
we
dev
e
l
o
pe
d, i
n
cl
u
d
i
n
g
m
a
xim
u
m
speed,
hi
g
h
est
al
t
i
t
ude, m
a
xim
u
m
ascendi
n
g
s
p
eed
, pay
l
oa
d,
and
bat
t
e
ry
ru
nt
im
e. The m
e
asur
e
d
perform
a
nce re
sults are s
h
own in Ta
ble
4. All th
ese
m
easu
r
ed
fligh
t
p
e
rfo
r
man
ces satisfy o
u
r requ
irem
e
n
ts of
th
e h
e
lico
p
t
er
we
n
eed in
order to im
p
l
e
m
en
t a d
i
saster area m
o
n
ito
ring
syste
m
.
Table 4. Performance
Measure
m
ent
Re
su
lts o
f
th
e Hex
-
ro
t
o
r
Helicop
ter
Measure
m
ent It
e
m
Measured Va
lue
M
a
xim
u
m
speed
53 k
m
/h
Highest altitude
800
m
M
a
xim
u
m
ascending speed
5.
3
m
/
s
Pay
l
oad
1000
g
Batter
y
r
un ti
m
e
16
m
i
nutes
6.
CO
NCL
USI
O
N
In t
h
i
s
pa
per
,
we p
r
op
ose
d
a
di
sast
er area
m
oni
t
o
ri
ng sy
s
t
em
usi
ng el
ec
t
r
i
c
gr
ou
n
d
ve
hi
cl
es an
d
el
ect
ri
c hel
i
c
opt
ers.
We de
ve
l
ope
d a soft
wa
re fram
e
wor
k
f
o
r t
h
e
AR
.D
r
o
ne 2.
0 t
o
i
m
pl
em
ent
t
h
i
s
sy
st
em
. W
e
obt
ai
ne
d t
h
e fl
i
ght
dat
a
of t
h
e AR
.D
ro
ne 2.
0 an
d i
nves
t
i
g
at
ed t
h
e rel
a
t
i
onshi
p bet
w
ee
n
fl
i
ght
pa
ram
e
t
e
rs an
d
fl
i
ght
dat
a
.
We
fo
u
nd t
h
at
t
h
e
EAM
val
u
e i
s
t
h
e m
a
xim
u
m
bendi
ng
an
gl
e, t
h
at
t
h
e
pi
t
c
h v
a
l
u
e i
s
a pe
rce
n
t
a
ge
of t
h
e m
a
xim
u
m
ben
d
i
n
g an
g
l
e, an
d t
h
at
a
p
i
t
c
h val
u
e
o
f
0
.
1 i
s
ap
p
r
o
x
i
m
at
el
y
25%
of t
h
e EAM
.
T
h
e
re
sul
t
of
ou
r ex
peri
m
e
nt
pro
v
e
d
t
h
at
t
h
e i
d
eal
val
u
e
fo
r t
h
e EAM
i
s
0.
2, an
d t
h
at
val
i
d
pi
t
c
h
val
u
es are 0 t
o
0.
3.
We
found t
h
at the
flight s
p
eed
data in the na
vdata cannot
b
e
tru
s
ted.
W
e
fo
und
th
at th
e
actu
a
l targ
et-ben
d
i
n
g
fl
i
ght
pi
t
c
h a
n
g
l
e of
t
h
e
AR
.
D
ro
ne ca
n
be cal
cul
a
t
e
d
by
a
fo
rm
ul
a usi
n
g
t
h
e EAM
a
n
d
pi
t
c
h
perce
n
t
a
ge
.
We al
s
o
d
e
vel
ope
d a
he
x-
r
o
t
o
r
hel
i
c
o
p
t
e
r
usi
n
g t
h
e D
J
I
In
n
ovat
i
o
ns
fr
am
e ki
t
and
a
n
a
u
t
o
pi
l
o
t
sy
stem
nam
e
d
Wo
o
K
o
n
g
-M
t
o
satisfy
o
u
r
r
e
qui
rem
e
nts fo
r ou
r di
sast
er
a
r
ea
m
oni
t
o
ri
ng syste
m
. At present,
t
h
e he
x-
rot
o
r
h
e
l
i
c
opt
er i
s
o
p
e
rat
e
d
usi
n
g a
pr
o
p
o
r
t
i
onal
ra
di
o c
ont
rol
sy
s
t
em
. Our
next
st
ep i
s
t
o
de
vel
op t
h
e
e
m
b
e
d
d
e
d
syst
e
m
an
d
software th
at will en
ab
le flig
h
t
o
p
e
ratio
n
b
y
co
mp
u
t
ers and
to
i
n
v
e
stig
ate its
actu
a
l
fl
i
ght
m
ovem
e
nt
s si
m
i
l
a
rl
y
.
Th
is stud
y con
s
titu
tes a first step
to
w
a
rd realizin
g
a m
o
n
ito
rin
g
sy
ste
m
u
s
in
g
electric g
r
o
und
v
e
h
i
cles and
el
ectric h
e
lico
p
t
ers. Mu
ch
work
is left to
d
o
i
n
ord
e
r to
im
p
l
e
m
en
t th
is
m
o
n
ito
ri
n
g
system
.
Fo
r
exam
pl
e, we h
a
ve t
o
st
udy
h
o
w t
o
at
t
a
i
n
au
t
o
n
o
m
ous fl
i
g
h
t
by
com
put
er
cont
rol
,
h
o
w t
o
p
r
og
ram
a fl
ight
t
o
t
r
ack a
n
E
V
b
y
GPS,
an
d
w
h
at
ki
nd
o
f
i
m
age a
n
al
y
s
i
s
i
s
use
f
ul
t
o
fi
nd
vi
ct
im
s of
na
t
u
ral
di
sast
e
r
s.
The
ch
allen
g
e
n
o
w
is to
id
en
tify these po
ssi
b
ilitie
s.
ACKNOWLE
DGE
M
ENTS
Th
is
work was sup
ported
b
y
a JSPS Gran
t
-
in
-Aid
fo
r Scien
tific
Research,
Gran
t Nu
m
b
er
242
460
68.
We
wou
l
d like to
th
ank
Mr. H. Mach
i
n
aka, Mr. K.
Oki
m
ura, and Mr. K. Im
ai
for assistance with
the
measurem
ent of flight
data.
REFERE
NC
ES
[1]
K. Mase, “Information and Communicati
on Tech
nolog
y
and Electric
Vehi
cl
es
– P
a
ving the W
a
y
Towards
a S
m
art
Com
m
unity,”
IEICE Transaction
s
on Communica
tions
, vo
l. E95-
B, no. 6, pp. 190
2-1910, 2012
.
[2]
T. Ohm
i
nato,
et al.
, “Volcano
Observations using an Un
manne
d Autonomou
s Helicopter: S
e
ismic and GPS
Obs
e
rvations
Ne
ar th
e Ac
tive
S
u
m
m
it Area of S
a
kurajim
a
and Ki
ris
h
im
a Volcano
,
J
a
pan
,
”
EGU G
e
ne
ral
Asse
mbly
2012
, pp
. 8575
,
Apr. 2012.
[3]
Y. Matsuda and
K. Mase, “A
Proposal of
Ad
Hoc Networks Using Mini-EVs,”
Technical
Report of IEIC
E
,
AN2011-29, pp.
69-74, Oct.
2011
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
RA I
S
SN
:
208
9-4
8
5
6
Real
Fl
i
g
ht
M
o
vement
s
of
El
e
c
t
r
i
c
Hel
i
c
opt
e
rs f
o
r
Maki
ng
a
Di
sast
er
Are
a
M
o
ni
t
o
ri
n
g
…
(
T
akuy
a
S
a
i
t
o
)
83
[4]
T. Suzuki,
et al.
, “
R
eal-
tim
e Hazard M
a
p Gen
e
rat
i
on us
ing S
m
all Unm
a
nned Aerial Vehi
cle
,
”
Proc. Of SIC
E
Annual Con
f
erence
, pp. 443-446
, Aug. 2008.
[5]
K.
Alexis,
et
.
a
l
.
, “Coordination of Helicopter UAVs for
Aerial Forest-Fire Surveillance,
”
Applications of Intelligent
Control to
Engin
eering S
y
stems
,
vol. 39
, pp
. 169-
193, 2009
.
[6]
T. Saito, H. Machinak
a,
and
K. Mase, “A Proposal of a Disa
ster Area Monitor
i
ng S
y
stem,”
Technica
l Report
of
IEICE
, vol. 112(
239), pp
. 64-68
,
Oct. 2012
.
[7]
K. M
a
s
e
and
T.
S
a
ito, “
E
lec
t
ri
c-
Vehicl
e-Bas
e
d
Ad Hoc
Networ
king and Surveillance
for Disaster Recover
y
,”
Th
e
Ninth Internatio
nal Conference
on Networking a
nd Services (
I
CNS2013)
, pp. 87
-93, Mar
.
2013
.
[8]
Parrot, “AR.Dro
n
e 2.0
,
” av
ailable from http://a
rd
rone2.par
r
ot.co
m
/, a
ccessed
on
Sep. 06
, 2013
.
[9]
DJI Innovations, “DJI Innovations Flight Con
t
rol Exp
e
rts,” available from http:/
/www.dji-in
novations.com/,
accessed on Sep
.
06, 2013.
[10]
Parrot, “AR.Drone 2.0 develope
r guide SDK 2.0,” ARDRONE open
API platform, available fro
m
https://pro
jects.ardrone.org/
,
accessed on Sep. 06,
2013.
BIOGRAP
HI
ES OF
AUTH
ORS
Takuy
a Saito
r
ece
ived
his
B.
E.,
M
.
E.
,
and
Dr.
Eng
.
Degr
ees in Computer
Scien
c
e from
Waseda Univers
i
ty
, Tok
y
o, Japan, in
2003, 200
5, and
2011, respectiv
ely
.
In 2
011, he was a
research
assistan
t in
the Departm
e
nt of
Comput
er
Scien
ce
and
En
gineer
ing, Waseda University
.
Since 2012, he is an assistan
t professor in th
e Research Institute for Natur
a
l Hazards and
Disaster Recover
y
, Niig
ata Univ
ersity
. He h
a
s p
a
rticipated
in m
a
n
y
robot con
t
ests and he has
won man
y
aw
ar
ds in contests such as Open
-R
Techno
Forum, Robo-One, and
RoboCup. His
research
inter
e
sts include humanoid robot cont
r
o
l, movement and inte
ractiv
e communication,
im
age an
al
ys
is
,
augm
ented r
e
a
lit
y,
3D gr
aphics
,
computer music,
and unm
a
nne
d ae
ri
al
ve
hi
cl
e
s
.
Kenichi Mase
r
ece
ived th
e B.
E., M
.
E
.,
and Dr. E
ng. Degrees in Electr
i
cal Engineer
ing from
Waseda Univers
i
ty
,
Tok
y
o, Japan, in
1970, 1972
, and
1983,
resp
ectively
.
He join
ed Musashino
Electrical Communication
Lab
o
ra
tories of NTT Public Corpor
ation
in 1972
. H
e
was
Execu
tiv
e
Manager, Communications Q
u
ality
Labor
at
or
y, NTT
Te
lecom
m
uni
catio
ns Network
s
Laboratories fro
m
1994 to 1996 and Com
m
unications
Assessm
e
nt Labor
ator
y
,
N
TT Multim
edi
a
Networks Labor
atories from 1996 to 1998. He
join
ed Niigata
University
in 19
99 and is now
P
r
ofes
s
o
r Em
eritus
,
Niigat
a Univ
ers
i
t
y
, Ni
igat
a,
J
a
pan. H
e
re
ceiv
e
d IEICE Bes
t
P
a
per Award in
1994, the
Telecommunications Advanced Foundati
on Award in
1998, and Best Paper Award,
International Academ
y
,
Resear
ch, and Industr
y Association in
2013. His res
earch
interests
includ
e commu
nications network design and traffic
contro
l, quality
of service, mobile ad hoc
networks, and wireless mesh
ne
tworks. He was President of IEICE-CS in 2008 and Vice
President of
IEI
C
E in
2011
and
2012. Prof. Mase is
an I
EEE and
IEICE Fellow
.
Evaluation Warning : The document was created with Spire.PDF for Python.