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