1

TEG-0322 April 2007

4th Flex-TEG MTBASt, Bergisch Gladbach

Development of a Biofidelic Flexible Pedestrian Legform Impactor Type GT

(FLEX-GT)

Development of a Biofidelic Flexible Pedestrian Legform Impactor Type GT

(FLEX-GT)

Atsuhiro KonosuTakahiro Issiki

Japan Automobile Research InstituteMasaaki Tanahashi

Hideki SuzukiJapan Automobile Manufacturers Association, Inc.

Back ground

2

Back ground

• The flexible pedestrian legform impactor type GT prototype (Flex-GT-prototype (called as Flex-GTα in the previous report1), 2))) was developed in Spring 2006.

• In this version, 1) the range of motion of the knee region, 2) the light weight of the bone parts, as well as 3) the biofidelity under assembly level (Thigh-Knee-Leg connected level) are improved.

• However, a validation of the biofidelity under assembly level was not completely conducted, so it still needs to be validated.

• Thus, further validation study on the biofidelity of this impactor, detailed computer simulation analysis was performed.

Methodology

3

0

200

400

600

800

1000

1200

-400 -200 0 200 400 600 800

Horizontal (mm)

Ver

tical

(mm

)

H2

H3

H1

L1

L2

BLE

BP

SPBLE: Bonnet leading edgeBP: Bumper SP: Spoiler

Computer simulation models

0

200

400

600

800

1000

1200

1400

1600

1800

2000

-1600 -1200 -800 -400 0 400

Horizontal (mm)

Ver

tical

(m

m)

Impact Speed: 11.1 m/s

H I: 25 mm (base)

Human Model

Flex-GT-prototype

Model SimplifiedCar

models

H I: Impact height

Simplified Car Model ParametersParameter Unit Level 1 Level 2 Level 3K1 (BLE stiffness*) mm 0.4 0.6K2 (BP stiffness**) JC*** 0.7 0.8 1.0K3 (SP stiffness**) JC*** 0.6 0.8 1.0H1 (BLE height) mm 650 700 750H2 (BP height) mm 450 490 530H3 (SP height) mm 250 270 350L1 (BLE lead) mm 125 200 275L2 (SP lead) mm -20 0 30* Stiffness is changed by steel plate thickness.

** Stiffness is changed by joint characteristics.

*** JC: Joint characteristics

# BLE: Bonnet leading edge, BP: Bumper, SP: Spoiler

As for the Flex-GT-prototype model, Impact height (HI) base + 50 mm and base + 75 mm is also calculated.

Explanation of Simplified Car ModelOver view – oblique front projection drawing

BP elements

BLE elements

SP elements

BLE: Bonnet leading edgeBP: BumperSP: Spoiler

4

900 mm

BLE elements(Shell elements, Deformable)

SP elements(Shell elements, Rigid)

500 mm

YXZ

500 mmFixedpart

Fixed part

50 mm

BLE: Bonnet leading edgeBP: BumperSP: Spoiler

50 mm

BP elements(Shell elements, Rigid)

Explanation of Simplified Car ModelFrontal view of the car

YX

Z BLE: Bonnet leading edgeBP: BumperSP: Spoiler

BP joint(Y translation)

BP elements(Shell elements, Rigid)

R25100 mm50 mm

90 mm

BLE elements(Shell elements, Deformable)

SP joint(Y translation)

SP elements(Shell elements, Rigid)

R25100 mm50 mm

90 mm

150 mm

100 mm

50 mm

100 mm

Explanation of Simplified Car ModelSide view of the car

5

0

2

4

6

8

10

12

14

16

0 20 40 60 80 100 120

Displacement (mm)

For

ce (

kN)

JC 1.0JC 0.8JC 0.7JC 0.6

Explanation of Simplified Car ModelBP and/or SP joint properties of the simplified car model

Simplified Car model specifications

K1 K2 K3 H1 H2 H3 L1 L2(BLE stiffness*) (BP stiffness**) (SP stiffness**) (BLE height) (BP height) (SP height) (BLE lead) (SP lead)

mm JC*** JC*** mm mm mm mm mmS1 0.4 0.7 0.6 650 450 250 125 -20S2 0.4 0.7 0.8 700 490 270 200 0S3 0.4 0.7 1.0 750 530 350 275 30S4 0.4 0.8 0.6 650 490 270 275 30S5 0.4 0.8 0.8 700 530 350 125 -20S6 0.4 0.8 1.0 750 450 250 200 0S7 0.4 1.0 0.6 700 450 350 200 30S8 0.4 1.0 0.8 750 490 250 275 -20S9 0.4 1.0 1.0 650 530 270 125 0S10 0.6 0.7 0.6 750 530 270 200 -20S11 0.6 0.7 0.8 650 450 350 275 0S12 0.6 0.7 1.0 700 490 250 125 30S13 0.6 0.8 0.6 700 530 250 275 0S14 0.6 0.8 0.8 750 450 270 125 30S15 0.6 0.8 1.0 650 490 350 200 -20S16 0.6 1.0 0.6 750 490 350 125 0S17 0.6 1.0 0.8 650 530 250 200 30S18 0.6 1.0 1.0 700 450 270 275 -20

* Stiffness is changed by steel plate thickness.

** Stiffness is changed by joint characteristics.

*** JC: Joint characteristics

# BLE: Bonnet leading edge, BP: Bumper, SP: Spoiler

SimplifiedCar Model

ID

Specifications of the simplified car models (total 18 types)Based on design of experiment method, L18 orthogonal table is utilized

6

20°

Human model 3),4)

Simplified car model

a) Back view of HM b) Side view of HMHM: Human Model

Posture of the human model

Setting of the simplified car model for the human model

BP, SP for initial impact side of leg

BP, SP for opposite side of leg(non-initial impact side of leg)

7

Front placement position of the Flex-GT-prototype model to the simplified car model

Flex-GT-prototype model

Simplified car model

a) Side view of SCM b) Frontal view of SCMSCM: Simplified car model

Leg measurement(Tibia bending

moment)

Kneemeasurement

(MCL Elongation)

Flex-GT-prototypemodel

Human model

Measurement points of the human model and the Flex-GT-prototype model

Tibia bending moment of human model can be obtained directory

In this study, tibia bending moment of Flex-GT-prototype is estimated from each part of bone core strain using bone core 3 point bending test results not using conventional dynamic 3 point leg bending test like before (dynamic 3 point bending test include high inertia force of leg then difficult obtain appropriate conversion values).

8

Results

Results: Relationship between the Human model and Flex-GT prototype modelResults: Relationship between the Human model and Flex-GT prototype model

Flex-GT prototype modelHI: base

Flex-GT prototype modelHI: base + 50 mmTibia

Knee-MCL

Flex-GT prototype modelHI: base + 75 mm

• To lift up the impact height of the Flex-GT can be obtained better correlation to the human one.

y = 0.6775x + 97.876

0

100

200

300

400

500

0 100 200 300 400 500

Human Model, Tibia Bending Moment, Max.-abs.- (Nm)

Fle

x-G

T-p

roto

type

Mod

elT

ibia

Ben

ding

Mom

ent (

RC

AL)

Max

.-abs

.- (N

m)

(Rtibia = 0.73)

y = 0.4957x + 15.939

0

5

10

15

20

25

30

35

40

45

50

0 5 10 15 20 25 30 35 40 45 50

Human Model, Knee MCL Elongation, Max. (mm)

Fle

x-G

T-p

roto

type

Mod

elK

nee

MC

L E

long

atio

nM

ax. (

mm

)

(RMCL = 0.40)

y = 0.9977x - 12.325

0

50

100

150

200

250

300

350

400

450

500

0 100 200 300 400 500

Human Model, Tibia Bending Moment, Max.-abs.- (Nm)

Fle

x-G

T-p

roto

type

Mod

elT

ibia

Ben

ding

Mom

ent (

RC

AL)

Max

.-abs

.- (N

m)

(Rtibia = 0.90)

y = 0.6924x + 8.0156

0

5

10

15

20

25

30

35

40

45

50

0 5 10 15 20 25 30 35 40 45 50

Human Model, Knee MCL Elongation, Max. (mm)

Fle

x-G

T p

roto

type

Mod

elK

nee

MC

L E

long

atio

nM

ax. (

mm

)

(RMCL = 0.66)

y = 0.9913x - 13.673

0

100

200

300

400

500

600

0 100 200 300 400 500 600

Human Model, Tibia Bending Moment, Max.-abs.- (Nm)

Fle

x-G

T-p

roto

type

Mod

elT

ibia

Ben

ding

Mom

ent (

RC

AL)

Max

.-ab

s.- (

Nm

)

(Rtibia = 0.82)

y = 0.9119x + 2.0157

0

5

10

15

20

25

30

35

40

45

50

0 5 10 15 20 25 30 35 40 45 50

Human Model, Knee MCL Elongation, Max. (mm)

Fle

x-G

T-p

roto

type

Mod

elK

nee

MC

L E

long

atio

nM

ax. (

mm

)

(RMCL = 0.83)

9

Discussions

Upper body tend to stay at the initial position because of its high inertia, as a result, upper body tend to lift up the knee joint and leg positions vertically. Impactor lift up motions of

knee joint and leg position are relatively small .

b) Flex-GT-prototype model, HI: basea) Human model, HI: base

Upper body effect (1): Lifting up the lower limb

-10

0

10

20

30

40

50

60

70

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 Ave.

Simplified Car Model ID

Kne

e jo

int v

ertic

al p

ositi

on-

from

its

initi

al p

ositi

on -

(mm

)

Human ModelHI: base

Flex-GT-prototype ModelHI: base

HI: Impact height

0

20

40

60

80

100

120

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 Ave.

Simplified Car Model ID

Kne

e jo

int v

ertic

al p

ositi

on-

from

its

initi

al p

ositi

on -

(m

m)

Human ModelHI: base

Flex-GT-prototype ModelHI: base

HI: Impact height

a) Tibia bending moment maximum timing

b) Knee MCL elongation maximum timing

10

Upper body effect (2): Inhibition of thigh behavior

The thigh can move easily compare to the human one, as a results, tend to generate large bending angle at the knee joint position.

b) Flex-GT-prototype model, HI: basea) Human model, HI: base

Upper body tend to stay at the initial position because of its high inertia, as a result, the thigh behavior is disturbed.

Thigh angle46.4°

Leg angle29.6°

S11 (30 ms) In order to sift up the impact height, leg rotation is facilitated, as a results, the knee bending angle can be comparable to the human one.

S11 (30 ms)

Thigh angle31.2°

Leg angle10.7°

Thigh angle48.5°

Leg angle8.6°

S11 (30 ms)

c) Flex-GT-prototype model, HI: base +75mm

-60

-40-20

020

40

6080

100

0 10 20 30 40 50 60

Time (ms)

Ang

le (d

eg)

Flex-GT prototype modelHI: base + 75 mm

-60

-40-20

0

20

4060

80

100

0 10 20 30 40 50 60

Time (ms)

Ang

le (d

eg)

Flex-GT prototype modelHI: base

-60-40-20

020

40

6080

100

0 10 20 30 40 50 60

Time (ms)

Ang

le (d

eg)

Human modelHI: base

Thigh angle46.4°

Leg angle29.6°

Thigh angle31.2°

Leg angle10.7°

Thigh angle48.5°

Leg angle8.6°

Thigh angleLeg angleKnee angle (Thigh angle - Leg angle)

Flex-GT-prototype ModelHI: base

Flex-GT-prototype ModelHI: base + 75 mm

Human ModelHI: base

S11 (30 ms) S11 (30 ms) S11 (30 ms)

S11

Upper body effect (2): Angle of the thigh, leg, and knee (S11, example)

knee angle is bigger than the human one

knee angle is comparable to the human one

Knee angle of human model

11

• From the results of this analysis, an impact height 50-75 mm higher than the base (25 mm) more correlated with the human model. One of the reasons for this is seemed as to be the effect of the presence or absence of the human upper body.

• The human upper body has great inertia force because of its size in mass relative to the leg, which tends to stay relatively at the initial position even after the leg crashes into a car. Therefore, during impact with a car, the upper body tends to lift up the leg overall.

• It is highly possible that these differences cause the difference in the loading condition on the tibia and the knee medial collateral ligament, and it is suggested that changing the impact height of the pedestrian legform impactor have effects to correct these differences.

• Moreover, the human upper body has the effect of inhibiting thigh movement due to its great inertia force.

• As mentioned above, the human upper body has great inertia force because of its size in mass relative to the leg, which tends to stay relatively at the initial position even after the leg crashes into a car. Therefore, during impact into a car, it inhibits thigh behavior to prevent the thigh from falling against the car.

• It is considered that shifting the impact position of the pedestrian legform impactor upwards especially facilitates rotation of the leg region of the pedestrian legform impactor, and as a result, the load occurring on the knee part has the same effect as in the human body.

Discussions and Conclusions

• Additionally, it has a chance that the difference in distribution of mass between the human body and the pedestrian legform impactor, while in the human body the bone part is very light in weight and a flesh part covers most of the mass, affects to the human and impactor differences.

• In the regulatory purpose pedestrian legform impactor, it is dif ficult to reduce the mass of the bone part to be equivalent to that of the human body because of various limitations such as incorporation of measuring censors, endurance, and testability.

• Moreover, to change the current impactor specification has a high risk for the developments itself (unexpected issue will be happened, that’s from our a lot of experiments).

• To keep the current specification of the impactor and to select best impact heights is therefore one of a good practical method we believe.

Discussions and Conclusions, contd.

12

References1) UN/ECE/WP29/GRSP/INF-GR-PS/Flex-TEG: Information on the Flexible

Pedestrian Legform Impactor GT Alpha (Flex-GT-alpha), TEG-021 (2006)2) UN/ECE/WP29/GRSP/INF-GR-PS/Flex-TEG: Evaluation Activities on Injury

Assessment Ability of the Flexible Pedestrian Legform Impactor GT Alpha (Flex-GT-alpha), TEG-022 (2006)

3) Takahashi Y., Kikuchi Y., Mori F., and Konosu A.: Advanced FE Lower Limb Model for Pedestrians, Proc. 18th International Technical Conference on the Enhanced Safety of Vehicles, Paper Number 218 (2003)

4) Takahashi Y., Kikuchi Y., and Mori F.: Development of a Finite Element Model for the Pedestrian Pelvis and Lower Limb, Society of Automotive Engineers World Congress, SAE paper No. 2006-01-0683 (2006)

Thank you for your attentions!