Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
PEDESTRIAN HEAD IMPACT CONDITIONS DEPENDING ON THE VEHICLE
FRONT SHAPE AND ITS CONSTRUCTION - FULL MODEL SIMULATION
ABSTRACT
Yutaka Okamoto, Tomiji Sugimoto, Koji Enomoto,
HONDA R&D CO., LTD.
Junichi Kikuchi
PSG CO., LTD.
The recent pedestrian accident data show the injury source for head changes from former ones. The
head contact points to the vehicle and contact conditions are thought to be influenced by the vehicle
front shape, its construction (rigidity) and pedestrian size. In this study, 32 simulation models are
calculated and the füll dummy , developing in HONDA, sied tests are conducted. The result shows
that not only vehicle front shape but also its rigidity much influences the head contact point and the
impact conditions are influenced by the pedestrian height.
key words: ACCIDENT ANALYSIS, EEVC, PEDESTRIANS, SIMULATION, SLED TESTS
FOR THE EVALUATION of Pedestrian Protection in Europe, the European Enhanced Vehicle
Committee (EEVC) - WG 1 7-report[ 4] is mainly used now. For the evaluation of pedestrian head
injuries, the report only takes the hood area of the vehicle into account. But from the recent Pedestrian
Accident Research it was found that, the windshield has a bigger effect on head injuries than the hood
area. One of the reasons for this phenomenon is the change ofthe front shape in recent vehicle models.
The head contact point and contact conditions are thought to be influenced by the construction
(rigidity), too. So in this study the Full Model Simulation is conducted to analyze the correlation
between the contact points, contact conditions and the front shape, its rigidity. The pedestrian models
are impacted not only at the center of the vehicle, like in many former studies, but also at the outer
side of the vehicle where the construction is different from the center part. Four different size
pedestrian models, two kinds of vehicle 3-D models and four impact positions are simulated. A full
scale dummy test is conducted to confirm the results of the simulation. In this test the pedestrian
dummy which HONDA R&D is developing, is used. This study shows that the head contact point is
much influenced by the front construction (rigidity) of the vehicle and that the head contact point is
more backward than the Wrap Around Distance (W.A.D.) which is measured from the front shape of
the vehicle. Furthermore the impact condition, head velocity and angle, are influenced by the
pedestrian size.
NHTSA CONDUCTED a Pedestrian Injury Causation Study (PICS) from 1977 to 1980. They were
interested in the influence of the vehicle front shape of recent models, and conducted another
pedestrian accident research, Pedestrian Crash Data Study (PCDS) from 1994 to 1 998 where only
recent model year vehicles were inspected [5]. Kristie L. Jarrett et al ( 1 998) [6] analyzed the data from
IRCOBI Conference - Montpellier (France), September 2000 281
both studies and the items found to be influenced by the change of front shape are the vehicle -pedestrian interaction, the increase in injury because of the windshield and A-pillar, and the decrease
of thorax, abdomen and pelvis injuries. But it must be remembered that the vehicle speed of AIS 5&6
injury cases on windshield or A-pillar is more than 40km/h. Under 40km/h, the hood is stil l the main
injury source for those cases. Dietmar Otte ( 1 999) [3] indicated similar results using the accident data
of Hannover Medical University. NHTSA released the PCDS data and Fig 1 , Fig.2 shows one example
of the analysis.
In PCDS they try to interview the drivers or pedestrians and investigate the pedestrian orientation to
100%
80%
20%
lnjury Body Region on AIS 5&6
PICS
....------,100% • Hcad.Ncck.Facc
O Chcst 80"1.
8 Abdomen
OSpinc 6-0%
OPcMc·Hip
40"
20%
PICS PCDS PCDS
Fig. 1 Injury Body Region and Injury Part Distribution for PICS and PCDS
(•o) H.O
)0.0
H.O
20 0
I S.O
10.0
s.o
o.o .,:„JS J6. 40
Sowf ct:PCDS
.1 1 . 0 l m ptcl Vclocil)' (km/h)
OCowl Aru
OHood
QHood Ed1c 1
Fig.2 Distribution oflnjury Source for Head Injury (AIS 5&6) by Impact Velocity
the vehicle at the accidents (Fig. 3) . From these data the pedestrian orientation was determined for use
in the simulation model in this study.
SIMULATION MODEL
PEDESTRIAN MODEL - HONDA R&D developed the Pedestrian Create program (HONDA
Pedestrian Dummy Creator) [8]. The origin of this program is GEBOD (GEnerator of BODy), Joints
properties data are improved by using the former PMHS (Post Mortem Human Subject) test results
and are transferred to PAM-CRASH data format. This model is constituted by 1 5 segments and 14
joints. All segments are made of rigid body and have mesh on the surfaces. The joint properties have
the same value for each model size. Here the joint properties for child is said to be different from adult
ones, softer, but there are few child PMHS test data for model validation. So in this study adult joint
282 IR CO BI Confere11ce - Montpellier (France), September 2000
l�""fLoll
c ....... . ,,,,.,.,, ,�.„„
lh•flC•'"'"''"''"„
,„„,...-...v.11o.1o
Distribution of Vehicle Movement
• • ' Tuming Left ' ' .... Tuming Right
• „ •
1 0 )0
Goin� Straight
40 '" 60 10 ., •> Distribution of Arm Position •>
10
Distribution of Chest Orientation
facit\lVctick F�AWI)' l.dtSid::to RigltSiJelo Okt Vcticlc Vchiclc
Distribution of Leg Position 40 ����������������- •o�����������������
""""' 11 ...... 11 ..... t..-i c.._ „ t...... "'"""' H.ou,, 11ar.,. o.... c„"'.., 11„ , ... ,... u,......i hh _,,,,, ... """""'• �· �·.., ........ „ ... ... h„� ... _ 1-..
�' -.._ . -Togc1htt Aptl'I· Ap.n· A„n.L.cn �n· WO Fool Rialrlt foOI Bolti f'ut Ot•n Uob„••
l.tiully Riilltl.cs Lei fa,..1t4 ofT1k offtk 01T1hc Forw11d fo,..ud Lt"t Gro1o114 Gtou-4 Gtoo.cl
Fig.3 Yehicle Movement and Pedestrian Orientation at Pre-Crash
properties was taken for child model, too.
Accordingly the results should be considered
for only qualitative one (trend). Four kinds of
pedestrian sizes (Fig. 4) are used in this
study. The friction between foot and ground
is neglected because in the full-scale dummy
test the dummy is hanging at the time of
contact. In the simulation the pedestrian
model is put as close as possible to the
vehicle making sure that initial penetration
does not occur. Therefore the dropping
distance by the gravity before the crash is
negligible.
Pedestrian
Size
Height(m)
Weight(Kgf)
Similar Dumm
Large Male
1.87
99.0
AM95
( 'Y' . ··J
t\ :l i 1 ""
Average Small Male Female
1.76 1.52
78.0 46.0
AMSO AF05
Child
1 . 1 4
23.0
C6Y
Fig.4 Pedestrian Model Size on Simulation
VEHICLE MODEL - Two kinds of vehicle models with different hood edge height are used, a
passenger car and a utility vehicle. They are the 3D half body models based on the mass production
vehicles. The vehicle weight is equal to the catalogue value and the added mass is put at the aft end for
supplementing the shortage in the model weight. The vehicle parts in the frontal area are made in
detail but in the engine room only the engine is modeled and other parts have only weight at their
center of gravity. Vehicle speed is 40km/h and braking after the crash is not considered.
IRCOBI Conference - Montpellier (France), September 2000 283
Head Light
1 1 1 1
( 78 )j( 58 ) ( 2.58 ) ( 08 )
. . . � „ „„ „„ „, 200 1 1250 1 250 1
Front 8ulkhead
8umper 8eam
Impact Position (Standing Point)
Fig.5 Front Construction and Pedestrian Standing Points
Fig.6 Overall View of Calculation Model
CALCULA TION MODEL - In this study, head contact points and impact conditions are considered.
Items influencing them are not only the front shape but also its construction (rigidity), so the
pedestrians are impacted at the outer position too. Fig. 5 shows the impact positions with the lateral
measurements indicating the distance between the pelvis center of gravity and the vehicle center line.
The pedestrian orientation at pre-crash is decided from the PCDS analysis, thorax and face is lateral to
the vehicle, legs are fore and aft and arms are fore and aft at the side. Fig. 6 shows one of the
calculation models. Totally 32 cases are calculated in this study.
RESULTS OF SIMULATION
HEAD CONTACT POINTS are decided
through the animation. For the head
contact velocity, first a time analysis of
the head tri-axial combined velocity at the
center of gravity is made determining the
timing when the velocity changes rapidly.
The head velocity is decided as the value
at the above timing. For the head contact
angle, the head trajectory curve is
established and it is defined as dz/dx at
the contact timing (with z the vertical
displacement and x the horizontal
displacement). Fig. 7 shows the head
contact points for each condition. lt
shows that for the center area of the
vehicle, contact points are parallel to the
284
Passenger Car Utility Vehicle
• • �-- W.A.D. 1500
+ Large Male • Small Female
• Average Male • Child
Fig. 7 Head Contact Points
IR CO BI Conference - Montpellier (Fra11ce), September 2000
V clocity(km/h) 60
Passenger Car Vclocitvlkm/h\ 60
Utility Vehiclc
40�=:==�.= .. . = . . = . . . � .• �.� .. = . . = .. . = . . � . . � .. :::::-j�t-�-�-:--��;�:�-�.�������.�J:�
1 1
1 1 1
- · -· - · - ·i- · - · - · - . - . -· -· - · � · - · - · - · - · - · - · - · · ··· · · · -
1 - . . . . -� . · · . . . - ·-
20
OB 2.5B
1
5B 7B Impact Position
--+- - · Large Male
__._ Average Male
· - - - . . - . . . · ·• . . . - - ··· 20
OB 2.5B
„ .• „ ••.•• „.„ .• Small Female
-·-„·-· Child
Fig. 8 Head Contact Velocity by Impact Position
5B
Angle (dcg) Passenger Car Angle (dcg) Utility Vehicle 100 .-------,-------.-------..., 100
1 -.. ••• 1
80
Impact Position
7B
80
l . . . . · · · · · . . .. . . . . ·.�-.·:�.·-·:.-.-::·.·� �:� :�·. ::·;: :.: c:.·.·�.-„, ... ; . . - • -
---- ---- ·-+��.� .. .. . . �.�.�-�------t-------- -.�-�.�.�� . 60
40
20
OB
1
� 60
- - - - - - --
- - - -� -- - -- - -
- -� 1
-- - 40 . . . . . . .
· - � . . .
2.5B
20
5B 7B OB Impact Position
1
2.5B
- -+--· Large Male
__._ Average Male
•• „„„ .•• „ .•• „ Small Female
-·-„·-· Child
Fig. 9 Head Contact Angle by Impact Position
·····.I .
5B 7B Impact Position
front shape, but for outer side impacts they are more backward than the W.A.D. which is dependent
from the front shape. This phenomenon is found on each pedestrian size and vehicle type, and for the
passenger car it is remarkable. Fig. 8 shows the head contact velocity and in this figure the abscissa
means the pedestrian impact position. The head velocity of the child in all positions is lower than the
vehicle speed. This result is similar to Wismans et al ( 1999) [7] result and lower than the velocity in
the EEVC report. lt is increasing along the pedestrian height but the head velocity of the !arge male is
the same as that for the average male. Fig. 9 shows the head contact angle. The bigger the pedestrian
size, the smaller the angle is, for the passenger
car. This result is not in line with the EEVC
report. On the other hand for the utility vehicle,
the angle of the child head is the smallest with
the angle for the other pedestrian sizes being
similar to each other.
RESULT OF FULL DUMMY TEST
I� n
n
OB
4B
SB
6B
Passenger Car Utility Vehicle
Fore Aft Fore Aft
() () 0 () 0
() () 0 () ()
Table I SLED Test Condition
IRCOBI Conference - Montpellier (France), September 2000 285
Impact Position : OB Impact Position : 6B
Fig. 1 0 Standing Posture at Pre-Crash
Passenger Car Utility Vebicle
• • • D
• • •o
e Simulation • Test 0 1 est (Left Leg:AFT) (Left Leg:AFT) (Left Leg: FORE)
Fig. 1 1 Head Contact Points (Simulation & TEST)
HONDA R&D IS since 1997, developing a pedestrian dummy for pedestrian protection [ 1 ] , [2]. In this
study the Phase 1 dummy is used. The goal of Phase 1 dummy was to make the motion similar to
PMHS. So this dummy is enough for this study to confirm the head contact points and conditions.
Dummy size is l .76m height, 74.9kgf weight thus being similar to the average male size. The full
scale dummy sied test is conducted at Japan Automobile Research Institute (JARi) to check the simulation results. Dummy impact points are center, 500mm and 600mm from the center of the
vehicle. And to confirm the influence of the leg position the test is conducted both left leg forward and
aft at center and SOOmm position. Table 1 shows the test condition and Fig. 1 0 shows the dummy
286
Velocity (km/h) 55
Passenger Car
1 1
1 1 1 1 1 1 1
50 - - -
, - - - "T - - - r - - -, - - - T - - - 1- - - ,
- - -
45 - - - -, - - -
1 ' 1 ' 1
'
40
-
- -:- - - � -
- - � -- � - - - �
- _
· :.� - - � - - - 1 1 1 1 1 1 1 1 1 1 : 1 1 1 1 t 35 - - -: - - - : - - - :- - - -: - - - : - - - :- - - : - - - 1 1 1 1 1 1 1 1
Velocity (km/h) 55
Utility Vehicle
1 1 1 1
50 - - -, - - - T - - - 1- - - , - - - r - - -, - - - "T - - -
45
1 1 1 1
40 ____ --.-, ------T---
--�- :- - - � - - -:>.- :-_· � - - - � - --
' „/ 1 1
· - · ....1 . _ _ _ t 1 1 / ! 1 1
35 - - - : - - - �-:_. �: �. ::-� -f � - - � - - -: - - - � - - -
OB 2B 4B 6B 8B OB 2B 4B 6B 8B Impact Position Impact Position
-..-- Simulation -·•·- Test
Fig.12 Head Contact Velocity (Simulation & TEST)
Angle (deg) Passenger Car so �---�----=-�-�-�--,...----, 1 1 1 1 1 _ _ _ ! - - - J. - - _ L _ _ _ 1 _ - - .1 _ - _ L _ _ _ t - - -
1 1 1 70 - - -, - - - ; - - - r - - - , - - - ; - - - r - - -, - - -
1 1 1 1 1 1 - - � - - - • - - - L - - � - - - • - - - L - - � - - -
--�--�e...._' 1
�i- -�--- �---r-�- - - � -
-- �- -�---
1 1 1 1 ! - - � - - - • - - - L - - � - - - · - - L - - � - - -
50 ·--���� �-�1 ·--:--·��-�-�·-T:.�:�.� - -: - - -
- - _1 _ - - J. - - - L - - _1 _ - - .l - - - L - - - - - -
Angle (deg) 80
Utility Vehicle - - -1 - - - L - - - 1 - - - .L - - - '- - - j
-- - l- - -
1 1 : : : ./ : : 70 - - -, - - - f - - -, - - T - - - ,-/- - 1 - - - ,- - -
1 1 1 • .if „ ·--... -� „ .-_-: �.:. :-_-::.: .-� � _;. ,/ .!' -=',t:_-:_:-_:-::-:-.--
1 1 1 1 1 1
60 - - -, - - - 1 - - - , - - - 1 - - - - - - 1 - - - ,- - -- - -1 - - - .... - - - 1 - - - + - - - ,_ - - � - - - � - -
1 ' 50 - - -: - - - !. - - _, _ - - f - - -
,_ - - � - - - :- - -
- - -t - - - ... - - -1 - - - .,. - - - 1- - - „ - - - t- - -1
OB 2B 4B 6B 8B OB 2B 4B 68 8B Impact Position Impact Position
-e-- Simulation - ·•·- Test
Fig. 13 Head Contact Angle (Simulation & TEST)
IRCOBI Con/erence - Montpellier (France), September 2000
Oms
30ms
60ms
90ms
120ms
150ms
Impact Position : OB Impact Position : 5B
Fig. 14 Mode Comparison between Simulation and Sied Test for passenger car ( Simulation Model : Average Male Pedestrian )
IR CO BI Conference - Montpellier (France), September 2000 287
setting photos. Sied velocity is 40km/h. Fig. 1 1 , 1 2 and 13 show the resulting, head contact point,
contact velocity, and contact angle. Head contact point is more backward than the W.A.D. at the outer
side of the vehicle, which is similar to the result of the simulation. Left leg position makes a little
difference for the torsion of the dummy motion but the contact points and conditions are similar.
DISCUSSION
- 1 .FIG. 14 SHOWS the mode of simulation for the average male model and the full-scale dummy test result. The mode is similar for both the center position and outer position except for the leg motion.
Fig. 1 1 , 1 2 and 1 3 show the head contact, head velocity and contact angle respectively. The simulation
model is sufficient to analyze the head impact conditions.
-2.THE REASON that the head contact points at the outer position are more backward than the
W.A.D., is the influence of the construction (rigidity). When the front construction is weak, the
pedestrian leg goes into the vehicle before falling down and the falling point is not the front face but
the inside ofthe vehicle.
-3.THE HEAD CONTACT velocity depends on the pedestrian height. Especially the child head
contact velocity is lower than the vehicle speed. The head velocity depends on the pedestrian rotation
after the impact. When the head falls towards the vehicle, the head velocity will be higher than vehicle
speed and when the head transfers in the vehicle direction of moving, it will be smaller. The
pedestrian head rotation mode is decided by the input force height, the pedestrian center of gravity
(P.C.G.) and the pedestrian head center of gravity (H.C.G.). A simple rigid bar model is used to
confirrn this mode like Fig. 1 5 .
y
X
Lt : Overall Height (m)
Lh : Head Height(m)
Lg : Height of Pedestrian Center of Gravity (P.C.G.) (m)
Lf: Input Force Height (m)
s : Distance between P.C.G. and Input Force Point (m)
s' : Distance between P.C.G. and Head Center ofGravity (H.C.G.) (m)
F : Input Force (KN)
Vp : Horizontal Velocity of P.C.G. (m/sec)
Vh : Horizontal Velocity ofH.C.G. (m/sec)
iP : Angular Velocity around P.C.G. (rad/sec)
J : Moment oflnertia around P.C.G. ( kgmsec2 )
Lf W : Weight (Kgf)
Fig. 15 Simple Rigid Model
The motion equations around P.C.G. for the rigid bar model are W d2x dw F = -- - - - - - - - -(1) T = J- - - - - - - - - - (2) g d2t dt
From ( 1 ) and (2), P.C.G. horizontal velocity(Vp) and angular velocity(w) is
288
g V (t) = -Ft - - - - - - - -(3) p w
T w(t) = - t - - - - - - - - - -( 4) J
IRCOBI Conference - Montpellier (France), September 2000
From (3) and (4), the dummy head horizontal velocity(Vh) is
Vh(t) = V/t) - s' cv(t) g ss'
= (- - -)Ft - - - - - -(5) w J
(5) shows that head velocity and direction is found by
g ss' K =
W - 7 - - - - - - - -(6)
When K is 0, then the s' is the imaginary rotation center (I.R.C.). If H.C.G. is above I.R.C., the head velocity is higher than the vehicle speed and if H.C.G. is below I.R.C., the head velocity is lower. For
this model, the moment of inertia (J) is determined only by height (Lt) and weight (W)
W Lt2 J = -- - - - - - - - - - (7)
g 1 2
In (6), W and s' will concern with Lt and s will concern with Lt and vehicle bumper height. The
correlation between them is found for the simulation models in this study. Table2 shows the P.C.G.
and H.C.G. for each pedestrian model. Fig. I6 shows the correlation between Lt and s, s' and Fig. I 7 shows between Lt and W. The following approximates are adopted which are based on Lt.
From (7) to ( I I ), (6) can be expressed by Lt only. Fig. I 8 shows the correlation between Lt and K
when impacting at the center of vehicle. From these figures, the child head velocity is lower than the
vehicle speed and the head velocity is saturated around the average male height. This is similar to the
----------- Large Male Average Male Small Female Child
Height (Lt / cm) 187.0 1 76.0 152.0 1 14.0 Mass (W/ Kgt) 99.0 78.0 46.0 23.0
Height of Pedestrian Center of Gravity 103.6 96.3 81 .8 61 .4 (P.C.G. /cm)
Height ofHead Center ofGravity 172.9 162.4 140.1 101 .3 (H.C.G. /cm)
Table2. Dimension for each pedestrian model
s . s' (m) 0.8 Large 1 1 l 1 1 Average 1 Mal 1
- - !.. - - J. - - .J - - _i - - - ' - - - '- ·Male - 1.. - e .l - -1 1 1 1 Small 1 • 1 5 1 1 1 1 female 0.6 - - : - - ; - - : - - -: - - -
1 1 1 : ...... : - - i. - - � - - .: - _, _ _ -1- - - l- - - L- �„: - .- i..S' _
1 . 1 1 l 1 1 !.."" 1 1 1Ch1ld 1 1 1 ,„ ""1 1 ... • 1
0.4 - - T - i' - - ""i - - -, - - -1- - ... -/- - ,- - -
• .-.-:- - T - -1 1 1 1 ... J ...... 1 ) ....... • 1 1
- - „ - - .... - - � - - � - _,. -1 - - - 1- -„- ... - - 1- - - ... - -• 1 1 ... "(... 1 .,.,,,, .... • 1 1 1
02 _Pas�e:;_scr_ .!. _ ..,...� ... : ...
_ _ • _ _ -_!•·�·_ i _ _ - '- _ _
!.. _ _
.!. _ _ . ! -.... 1 1 ,,. . ... 1 1 1 1 1
1 ....... , ... . ,,.· 1 1 -Utility . - ..,. - �„.,_ ....... -::.· -j - - -1 - - -1- - - .... - - � - - + - -
Vchiclc . „ . !. · ...... 1 1 1 1 1 t 1
o ��--�-�-�-����-�-�� 1.0 1.2 1 .4 1.6 1.8
F ig. 16 Tue Relation between s, s' and Lt
2.0 Lt(m)
W(kgf) 120 Large 1
Male 100 - - ... - - + - - ... - - -4 - - -1 - - _,_ - - 1- - - ,_ - ... - -1 1 1 1 1 t Average 1 1 ' 1 1 1 1 1 Mal 1
80 - - L - - L - - � - - J - - -' - _ -'- _ _ ._: _ L _ _ L _ _ 1 1 1 1
60 - - i- - - + - - � - - -: -�=�lle - -:- - :- - - � - - T - -
1 1 1 1
40 - - r - - T - - , - - , - -1 - - - 1- - - i- - - ,_ - - r - -
1Child 1 ' 1 20 - - 1. - - J. - - .J - - .J - - _, - - - 1 - - - 1- - - L. - - - - -
1 1
o��-�-�-��������-�� 1.0 1.2 1.4 1.6 1.8 2.0
Lt(m) Fig. 17 The Relation between W and Lt
s' = -0.246Lt2 + I . 1 38Lt - 0.58 I - - - - - - - -(8)
s = O. I 43Lt2 + O. I 54Lt - 0.223 - - - - - - - - - (9) (for passenger car)
= 0. 1 43Lt2 + O. l 54Lt - 0.353 - - - _ - _ _ _ _ (10) (for utility vehicle)
W = I 25.98Lt2 + 275.68Lt + I 73.63 - - - - - -- (1 I)
IR CO BI Conference - Montpellier (France), September 2000 289
result of the simulation. Accordingly the head contact velocity for the child area of the vehicle is lower than the vehicle velocity, and that for the adult area is higher but the velocity for the average male
height is the maximum necessary for consideration.
K(ratio) Passenger Car (OB Standing) 0.6 .----,---.....--......... --......--�---,.--�
1 1 1 1 - - - , - - - - r - - - , - - - - r - - - , - - - - - - - -
' 1 1 1 1 1 1 0.4 - - - 1 - - - - ,- - - - "'i - - - - ,- - - - 1 - - - - , - - - -
1
- - - ..1 - - - 1 - - - ..! - - - - ·- - - - ! - - - _ l_ - - -1 1 1 1 Chü�
0.2 - - - � - - - - :- - - - � - - - - :- - - - � - - - - :- - - -' 1 '
- - - .J - - - - ,_ - - - .J - - - '- - - - .l - - - - · - - - -1 Smull 1 1
f----'---.1----'---�',..Ft""mal_•__,'A";.�;• 1 Lt(m)
0 0.75 1.0 1.25 l.1'----2.0-largt
�-�--�-�--�--�Malt
K(ratio) Utility Vehicle (OB Standing) 0.6
1 1 1 1 1 1 - - - "'i - - -'
- - - 1 - - - - ,- - - - "j' - - - - , - - - -
1 1 1 1 1 1
0.4 - - - � - - - _ :- _ _ _ shild _ _ _ :- _ _ _ � _ _ _ - :- _ _ _
1
- - - ..1 - - - - ·- - - - - - - _ ,_ - - - ..!. - - - _ 1 _ - - -1 1 1 1 1 1 1 1 1 1 1 1
0.2 - - - .J - - - - :- - - - � - - - :- - - - � - - - -·- - - -
1 1 1 1 Sma/I 1 _ _ _ ..J _ _ _ _ L _ _ _ .J _ _ _ _ _ FenUlle J. _
, _ _ _ _
1 Avnage 1 , Male 1 Lt m ol----'---.1----'----'-----"�--,r.==='""'I
0. 75 1.0 1.25 l.S 1.75 2.0 1 1 1 1 1 IArgt '---�--�-�--�-�--Malt
Fig. 18 Head Velocity Ratio(K) depending on the Pedestrian Height(Lt)
-4.THE HEAD CONTACT angle depends on the pedestrian height , too. Especially for the passenger
car, the child head contact angle is the largest for all pedestrian sizes. This result is different from the
EEVC test procedure. Fig. 1 9 shows the animation at the contact timing. The child is not carried over
to the vehicle and only the neck is bent. Furthermore the contact object is the hood which is nearly
horizontal and consequently the contact angle is large. On the other hand, the !arge male is carried
over and the whole body wraps around the hood . The contact object is the windshield which has an
initial angle so that the contact angle is small. For the utility vehicle, the child head contact angle is the
smallest. In this case the head collides directly with the hood edge before there is any neck bending
(Fig. 20). For other sizes, the contact object is below the windshield or on the hood and these areas are
nearly horizontal.
Child Pedestrian Large Male Pedestrian
Fig. 1 9 Head Contact for Passenger Car
Child Pedestrian Large Male Pedestrian
Fig.20 Head Contact for Utility Vehicle
290 IRCOBI Conference - Montpellier (France), September 2000
CONCLUSION
THE PEDESTRIAN FULL MODEL simulation is conducted to analyze the influence of the vehicle
front shape and its construction (rigidity) on the head impact conditions. The conclusions are summarized below.
- 1 .The head contact point is influenced by the front construction (rigidity). -2.The head contact velocity depends on the dummy height. The child head velocity is lower than the
vehicle speed and the taller the pedestrian, the higher his head velocity is. But this increased ratio is
saturating around that of the average male height. -3.The contact angle for passenger car depends on the pedestrian height and child head contact angle is the largest.
ACKNOWLEDGMENT
The authors would like to thank Dr. Roger Saul, NHTSA for his effort in the PCDS project and
making the data available. Also they would like to thank Ms. Majorire Saccoccio for getting the data,
too. Thanks to Mr. Kounosu , JARI for the sied test.
REFERENCES
[ 1 ] Akiyama A., Yoshida S.,Matsuhashi T.,Moss S.,Salloum M, Ishikawa H., Konosu A.
"Development ofHuman-Like Pedestrian Dummy", JSAE Spring Convention, 1999
[2] Akiyama A., Yoshida S.,Matsuhashi T.,Moss S.,Salloum M, lshikawa H., Konosu A.,
"Development of Simulation Model and Pedestrian Dummy", SAE paper No. 1999-01-0082,
1999
[3] Dietmar 0., "Severity and Mechanism of Head Impacts in Car to Pedestrian Accidents", IRCOBI,
1 999,pp. 329-341
[ 4 ] European Enhanced Vehicle-safety Committee, "Improved test methods to evaluate pedestrian
protection afforded by passenger cars", EEVC Working Group 1 7 Draft Report, 1 998
[5] Isenberg R.,Mavros S., "Update on the Pedestrian Crash Data Study, PCDS", ESV paper No. 98-
S6-0-05, 1 6th ESV, 1998, pp. 1 2 1 2- 1 223
[6] Jarrett K.,Saul R., "Pedestrian Injury,Analysis of the PCDS Field Collision Data", ESV paper No.
98-S6-0-04, 1 6th ESV, 1 998, 1 204- 1 2 1 1 [7] Wismans J., Happee R., "Pedestrian Protection Full-Body Simulations, Dummy Validation",
VDA Technical Congress, 1999, pp. 2 1 -29
[8] Yoshida S., Matsuhashi T., Matsuoka Y. "Simulation ofCar-Pedestrian Accident for Evaluate Car
Structure", ESV paper No. 98-Sl O-W- 1 8, 1 6th ESV, 1998, pp.2344-2348
IRCOBI Conference - Montpellier (France), September 2000 291