Accident Reconstruction Accident Reconstruction
• Goals of Accident Reconstruction • Data Collection / Accident
– What happened? Investigation
• Investigation of product/environment needs – Accident type, vehicles, injuries
– Data Collection – Gross motions of vehicles and
• Characterise/document crashes for data occupants
processing
• Reconstruction
– Litigation
• Assessing the cause of injury and collision – Calculate severity
cause – Detailed simulation of system dynamics
Organisations Terminology
• Crush : Quantity of Vehicle Damage
• Road Designers
• Yaw - Vehicle Rotation or Combination of
– Highway Department
rotation/forward motion
• Vehicle Designers • Delta (∆) V: Change of velocity
– Manufacturers, Safety Engineers • EBS, EES : Equivalent Barrier (Energy)
• Medical / Human Factors Researchers Speed
– doctors, ergonomics • PDOF : Principal Direction of Force
• Overlap
• Underride/Override
• Point of Impact (POI) / Point of Rest (POR)
Terminology Haddon’s Matrix
Direction Pre Crash Post
Crash Crash
Human
Crush
Overlap Vehicle
Location Environment
1
Example of Vehicle
Crash Parameters
Accelerations
Crash History Case 1
• Delta V: 30 20
Acceleration [g]
Velocity [m/s]
15
– is NOT the speed of impact
20
∆V 10
– described the speed change during
10
5
a crash phase 0
0 0.1 0.2 0.3
0
– has been a historical measure of 80
Crash History Case 2
15
accident severity
Acceleration [g]
"Crash" "Braking"
Velocity [m/s]
60
10
40 ∆V
5
20
0 0
0 0.05 0.1 0.15
Time [s]
Delta V Principal Direction of Force
• PDOF represents the line of
action for the force exerted on a
vehicle during a collision
V1
V2 PDOF
V1
V2
∆V
Why Delta V? Energy
• Historical - Unbelted Occupants • EBS / BEV / EES / ETS
– Correspondence to occupant – equivalent impact speed for a rigid
impact with vehicle interior barrier test
• Practical – Energy Absorbed in vehicle
– Can be easily calculated damage is expressed as an
equivalent speed
• Biomechanical - Response
E= 1/2 MV2
– acceleration * duration
V=EBS/BEV/EES/ETS
2
When are Delta-V and EBS Different?
Energy Equivalent Speed
1) When the crush is not equally distributed over the 2 collision partners
2) When the vehicle does not come to rest at the end of the crash phase
Scene Data Complete Reconstruction
• Requires information for all three
phases
• Critical information
– Point of Impact
– Point of Rest
Pre-Crash
• Response time = Perception-Decision Time +
Braking
Reaction Time
• Passenger Cars
• 85th percentile of drivers = 1.6s + 1.5s = 3.1s
• distance for response time – assume brakes are capable of locking all
wheels
– 100 km/h => 28 m/s => 87 m
• Braking and Steering: Driver Reactions – f=mu*g
– if not locked wheel, brake efficiency factor η
Pedal Action
activation
also known as lockup factor
Brake
• Trucks
Speed
– Brakes may not be capable of locking
Response wheels
time
Braking – long hills may produce overheating
– out of adjustment brakes
0 Time
3
Vehicle Handling/Vehicle
Tire Behaviour
Dynamics longitudínal slip 20%
Force
• What did the vehicles do prior to lateral slip angle 15 deg.
collision?
• Begin with tire-road interface locked / sliding wheel
– traction circle
X
Friction Force Long
Slip
Y ω α
longitudinal slip
Friction Force Lat
(v-r*w)/v v
Tires velocity
Yaw Critical Speed
• vehicle on threshold of • Speed to
spinning produce onset of
yaw
• maximum cornering
• Assume:
• critical cornering speed – constant speed
• upper bounds on vehicle speed – vehicle point ρ
mass
– constant friction
Vcrit = g ρ µ m
Yaw Analysis Example
ρ ρ´
Average radius
C 2m m
ρ= +
8 2
m
chord
4
Brake Marks on Site Example
• Slide to stop distance
– straight line, locked wheel braking 23 m
– road friction measured µ = 0.72
• Speed at start of skid?
Wbrake = ∆KE
Fbrake d brake = 1 / 2m(vi2 − v 2f )
µ mgd brake = 1 / 2mvi2
vi = 2 µ gd brake v = 18 m/s
ABS Brakes Post-crash
• similar to pre-crash BUT …
– tires disabled
– vehicle mass distribution changed
• spinout
– equivalent friction value
Spin Out Spin Equivalence
ω ,V Linear Velocity, V
Co Angular Velocity, ω
fG
Idealised
Fro
nt W
hee Actual
ls
Each wheel has:
- different speed Rea
rW
hee
- different distance ls Time
- different slip angle
µ eff = f ( µ ,V ,ω , K ) ⇒ µ eff < µ
5
Equations of Motion
• estimate effect for each wheel
r r
m&x& = ∑ F
r r r
Iα = ∑ F × r F
Yaw mark from a test
Yaw mark Skid mark
Crash Dynamics Application Restrictions
• Two Approaches • Momentum
– Momentum – Accurate scene evidence
– Energy – Vehicle Masses
• Options – preimpact speed
– FEM • Energy
– Lumped Mass Models – Vehicle Stiffness data
– limited crash speed range (20-70
km/h)
Impact
Examples
mvi 2
2
A kx 2
A E=
2
B mv f 2
2
v1
vi' ∆V
B
time
6
Crush Model Energy Approach
• Background
• Formal model of energy
– Determine severity of accident
from vehicle damage
dissipated in damaged vehicles
– discovered linear relationship
Force/width
between crush and vehicle impact G=Area
speed for barrier tests A2/(2B)
B
c
V
A
Residual Crush
Example
Energy Approach
• Renault • Rover
• Formulate model of energy A= 45400 N/m A= 67160
B=296900 N/m/m B= 870000
dissipated in damaged vehicles
G=3470 N/m C=2592
- linear force/deflection
A=> N/m, B=>N/m2
• Damage • Damage
Force/width
dF = Adx + Bdcdx C1=0.7 m C1=0.35
B
C2=0.44 m C2=0.2
dx
L=1.6 m L=1.67
A dF C1
w
Residual Crush L
dc
C2
Energy Calculation Accident Severity
E = Fd
• Equivalent Barrier Speed (EBS)
E = ∫ Ac( x) dx + ∫∫ Bc( x)dcdx
• Equivalent Energy Speed (EES)
B
E = A∫ c( x)dx + • Kinetic Energy change for
2∫
c( x) dx + G ∫ dx
2
vehicle is equal to energy
absorbed
Force/width
G=Area
mv 2
dF = EB
2
A /(2B)
w B 2
A 2 EB v=> EBS, EES
dc v=
Residual Crush m
7
Calculation of Energy Energy Calculation
(C1 + C 2) B Crush Measurements
E=A L + (C12 + C 22 ) L + GL
2 2 L2 L3 L4 L5 L6
E1 = 86200
C1
C2
E2 = 63300 C3
C4
2E
EES = C5
M C6
EES1 = 33 kmh EES 2 = 28 kmh
2( E1 + E2 ) 2( E1 + E2 )
E = A∫ dl + B ∫∫ dc dl
∆V1 = ∆V2 =
M M l lc
M 1 (1 + 1 ) M 2 (1 + 2 )
M2 M1
∆V1 = 47 kmh ∆V2 = 22 kmh Damage Damage
width Area
Crush Measurements Numerical Solution
E = A( width ) + B( area ) + G
2A
G=
B2
Area is numerical integration of crush measurements
Crush Coefficients Load Cell Walls
• Define Vehicle Stiffness, A & B Load cell walls can provide force-deflection information
for vehicles
• Defaults based on wheelbase
800
• Individidual vehicle values are
Force [kN]
Load Sensors 600
better 400
• Values are becoming available 200
for side and rear structures
0 200 400 600 800 1000
Displacement [mm]
8
Error Sources Error Sources
• Measurements - accuracy of field • Vehicle damage only reflects
measurements static crush, elastic rebound
• Stiffness Coefficients - not incorporated in formulation
– data specific to vehicles under
investigation
• Vehicle damage only partly
– application of rigid barrier tests to car- describes collision, pre & post
car crashes crash data needed for complete
– angled impacts may not deform vehicle damage
under similar conditions used to
generate stiffness data
Validation of CRASH Algorithm to
Crash Recorder Data (Default Stiffness Angle of Impacts
Data) Φ
Percent Error
Vehicle Measurement Actual Case Information
• Damage types • Problems in the field
– Direct – No scene data
– Induced – Missing / unavailable vehicles
– Complex impact configurations
– Measurement of vehicles
9
Crush Measurement
Crush Measurement Side
•Exterior Damage Profile
•Undamaged Exemplar Profile
•Crush Profile
Crush can not be measured directly
Side Impacts Software Tools
• CRASH 3
• WinCrash (CRASH)
– Damage and Momentum
Pole Impact Car-Car
Damage Analysis Fundamentals
1
Estimate
crush energy
of both vehicles 2
Estimate
weights
Ma and Mb 3
Calculate
Closing Speed
Optional 4
Estimate vehicle speeds Calculate vehicles’
from closing speed Delta-V
using additional info
10