2009-01-0092

Determining Vehicle Steering and Braking from Yaw Mark Striations

Gray Beauchamp, David Hessel, Nathan A. Rose, Stephen J. Fenton

Kineticorp, LLC

Tilo Voitel
Denver Police Department

Copyright © 2009 SAE International

ABSTRACT

This paper presents equations that relate the orientation

and spacing of yaw mark striations to the vehicle braking
and steering levels present at the time the striations were
deposited. These equations, thus, provide a link between
physical evidence deposited on a roadway during a crash
(the tire mark striations) and actions taken by the driver
during that crash (steering and braking inputs). This
paper also presents physical yaw tests during which
striated yaw marks were deposited. Analysis of these
tests is conducted to address the degree to which the
presented equations can be used to determine a driver’s
actual steering and braking inputs. As a result of this
testing and analysis, it was concluded that striated tire
marks can offer a meaningful glimpse into the steering
and braking behavior of the driver of a yawing vehicle. It
was also found that consideration of yaw striations allows
for the reconstruction of a vehicle’s post-impact yaw
motion from a single tire mark.

INTRODUCTION

Braking and steering will both affect the characteristics of

the striations that are sometimes present in the tire marks
that a yawing vehicle deposits. As an example, consider Figure 1 – A yaw mark deposited during testing
Figure 1, which is a photograph from testing conducted performed by Daily et al.
by Daily [1] that depicts a yaw mark deposited by a
vehicle during a counter-clockwise yaw. The vehicle that As will be explained in greater detail later, the striations in
deposited this tire mark was traveling into the page. a yaw mark indicate the travel direction of tire shoulder
Initially, the vehicle was unbraked, then, the brakes were blocks during the yaw. Since frictional forces act to
applied. As a result of this changing braking level, the oppose the direction of motion, these striations are also
striations in the tire mark undergo a change in direction. indicative of the direction of the frictional force acting on
these portions of the tire. In the case of increasing

The Engineering Meetings Board has approved this paper for publication. It has successfully completed SAE’s peer review process under the supervision of
the session organizer. This process requires a minimum of three (3) reviews by industry experts.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE.
ISSN 0148-7191
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of
the paper.
SAE Customer Service: Tel: 877-606-7323 (inside USA and Canada)
Tel: 724-776-4970 (outside USA)
Fax: 724-776-0790
Email: [email protected]
SAE Web Address: http://www.sae.org
Printed in USA
braking depicted in Figure 1, the direction of the forces on Except for in the cases of full lockup or 90-degree tire slip
the tire, and thus, the direction of the striations, transition angle, a tire rolls and slides as it deposits striations. This
from a direction perpendicular to the heading of the tire to sliding results in forces that deform the tire carcass
a direction parallel to the velocity direction of that tire hub laterally and load the leading shoulder of the tire (the
(tangent to the path of the tire mark). In other words, the shoulder is the area where the tread or bottom of the tire
no braking case references the angle of the tire and the and the tire sidewall meet). An extreme case of such
full braking case references the general direction of the shoulder loading is shown in Figure 2 [2]. When distinct
tiremark. Thus, the analysis of striations includes striations are visible in a yaw mark (Figure 1), it is the
consideration of the vehicle and the tire mark evidence. individual tread blocks on the shoulder of the tire that are
In most cases, the tire mark evidence by itself is responsible for the darker striations. The fact that the
insufficient to determine the amount of braking or steering shoulder tread blocks create distinct striations is
that was occurring. fundamental to the modeling presented below.

Of course, even in the absence of braking, tire mark

striations could make a similar transition to that shown in
Figure 1. In the absence of braking, the tire forces will be
oriented perpendicular to the tire heading, and thus, the
striations will also be oriented perpendicular to the tire
heading. In the case of steadily increasing slip angle, the
striations in a tire mark would transition in a manner
similar to that shown in Figure 1. The striations would
become parallel with the direction of the tire mark when
the slip angle reached 90 degrees. To take this a step
further, it should be noted that steering will influence a
tires slip angle, and thus, will also influence the
orientation of striations deposited by an unbraked tire.

The tire slip angle and braking level will both influence
the characteristics of the striations in a yaw mark. If one
considers only the orientation of the striations, then these
various influences will be indistinguishable. However, it
will be shown below that consideration of both the Figure 2 – Testing performed by NHTSA exhibiting tire
orientation and the spacing of the striations will allow the deformation
effects of slip angle and braking to be distinguished and
parsed out. Figure 3 is a schematic depicting the alignment of a tire
relative to a yaw mark that it is depositing. In Figure 3,
Thus, a relationship exists between the actions of a driver the contact patch, or portion of the tire in contact with the
and the characteristics of the tire marks their vehicle roadway surface, has been identified with a dark grey
deposits during a yaw. Formalizing this relationship can box. The tire mark is divided, with darker striations on the
enable reconstruction of a driver’s actions during a crash right being deposited by the shoulder blocks and lighter
based on physical evidence. The remainder of this paper striations on the left being deposited by the tread. Such a
presents and tests equations that relate the striation division in the darkness of the striations from left to right
characteristics (direction and spacing) to both the is visible in the tire mark in Figure 1. In Figure 3, the
percentage of braking slip and the slip angle of the tire at striations are neither perpendicular to the tire heading or
the time a yaw mark was deposited. parallel to the velocity direction, indicating that partial
braking is occurring. The dark striations from the
shoulder blocks will be the focus of the model to be
YAW MARKS
presented below. The slip angle, α, is labeled in Figure
1
3.
When a vehicle yaws, its velocity direction deviates from
its heading direction. This difference can result in tire
marks being deposited on the roadway, within which THEORY
striation marks are often visible, as discussed above. Tire
mark striations are direct physical evidence of the In this section, mathematical equations are presented
direction of the force that was applied to the tire at the that relate braking and steering to the orientation and
time the mark was deposited. Without braking, the tire spacing of striations deposited on the roadway. This
force is perpendicular to the tire heading, and thus, the theory assumes that the tire is rigid, which as shown in
striations produced are perpendicular to the tire heading. Figure 2, is obviously not the case. Later in this paper,
When a yawing vehicle locks its brakes, the resultant tire the implications of this rigid tire assumption are
force is in line with velocity direction of the tire hub, hence
the direction of the striations becomes in line with that 1 A full listing and description of the variables used in this paper can be found in

velocity direction. Appendix A.

2
discussed. The equations below also don’t consider the the tread blocks on the tire and is oriented such that it
fact that both the size of the shoulder tread blocks and connects two adjacent striations. This alignment orients
the spacing between them are variable around an actual the line TD parallel to the tire heading. Both triangles are
tire’s circumference. This is an important characteristic of completed by lines along the striation mark, as depicted.
tires to consider in practice and is discussed later. The striation angle, θ, is shown on the triangle. Because
TD is parallel to the tire heading, the angle Ψ is created in
the second triangle. In practice, the angle θ and the
distance SD can be measured from a striated tire mark.
The tread block spacing distance, TD, is a measurement
taken from the tire. Measurement techniques for θ, SD
and TD are discussed later.

Figure 3 – Tire depositing a striated tire mark.

TIRE SLIP ANGLE – Figure 4 depicts the same yawing

tire as Figure 3. It is the dark striations deposited by the
shoulder of the tire that form the foundation of the
modeling below, thus, these striations have been singled
out in Figure 4. The coordinate system depicted in Figure Figure 4 - Top view of a yawing tire depositing a striated
4 is consistent with the SAE tire axis system. It is right- tire mark.
handed, with the z-axis pointing into the page, and the x-
axis aligned with the tire’s heading. The angle α is the The triangles to the lower left of Figure 4 can be
tire’s slip angle, which is the angle between the tire’s geometrically related to obtain an expression for Ψ, given
heading and the tire’s velocity direction. The dark black by Equation (1) below.
lines represent the striations in the tire mark, deposited
by the shoulder treads of the tire, and they are oriented
 SD sin θ 
an angle Ψ from the tire’s heading. As drawn in Figure 4, Ψ = sin −1   (1)
both α and Ψ are positive. The angle θ is the angle  TD 
between the tire’s velocity direction and the direction of a Using Equation (1) along with geometric relationships
striation. The striations are a distance of SD apart in the between Ψ, α and θ, an expression for the slip angle, or
direction of the tire mark, or tire velocity direction. the angle of the tire relative to the tire velocity, can be
obtained, given below in Equation (2). All of the variables
In the lower portion of Figure 4, a zoomed in view of two in Equation (2) can be measured from either a tire or a
right triangles is shown. One side of both triangles is tire mark.
formed by a line connecting two adjacent striations in a
direction perpendicular to the striations. The hypotenuse  SD sin θ 
of the larger triangle is SD. The second right triangle has α = sin −1   −θ (2)
a hypotenuse of length TD and is denoted by a dashed  TD 
line. TD has a length equivalent to the spacing between
3
When multiple tire marks are deposited on the roadway, the tire at Positions 1 and 2. Between Positions 1 and 2,
vehicle yaw angles can be determined by aligning a the tire rotates β radians. While the tire travels from
vehicle with its tire marks. Using the equations above, the Position 1 to 2, it deposits a striation of length LS,
slip angles of each tire can also be determined. Rear tire depicted by the bold line in the top view image on the left.
slip angles are equal to the slip angle of the vehicle in the The tire mark has a width w. The tire slip angle and the
absence of tire misalignment or toe in/out design striation angle are indicated by α and θ respectively. The
differences. The difference between the heading direction contact patch length is also indicated on the diagram by
of the front tires and the heading direction of the vehicle the dashed line.
is the steering angle at the front tires. The angle of the
steering wheel can then be calculated through the
steering ratio.

BRAKING/SLIP PERCENTAGE – Along with the

orientation of the tire, the amount of braking of a yawing
tire is often of interest. Braking can be discussed in terms
of slip percentage, which can be calculated using some
of the parameters that appear in Figure 4. As will be
shown, slip percentage is dependant on fewer variables
than slip angle. This is important because in some cases,
slip percentage can be calculated when the slip angle
cannot.

Our discussion of slip will begin with the classic definition

of braking slip, which appears in several vehicle
dynamics texts. For instance, Gillespie states: “Slip of the
tire is defined by the ratio of slip velocity in the contact
patch to forward velocity” [3]. Mathematically, this is
shown in Equation (3) for the case of in-line braking.

V − ωr (3)
S=
V

In Equation (3), V is the velocity of a wheel hub, ω and r

are the rotational speed and radius of its tire,
respectively. The quantity ωr will be equal to V when the
tire is free rolling so the slip will be 0. Comparatively, at Figure 5 - Depositing of one striation mark. On the left, a
full wheel lockup when ω is equal to 0, slip will be equal to top down view. On the right, a right side view.
1.
Physically, the contact patch length, CPL, is the length of
Equation (3) assumes a tire that has its heading and its the portion of the tire that is in contact with the ground.
travel direction aligned. For a yawing vehicle, it is the Figure 5 depicts the contact patch length. In Position 1, a
velocity of the tire along its heading direction that is of point on the circumference of the tire marked with a black
interest. The velocity of the tire along its heading, VT, is dot has just come into contact with the ground at the front
the actual velocity multiplied by the cosine of the slip of the contact patch. While the striation is being
angle of the tire. Thus, Equation (5) gives braking slip for deposited, the tire rotates and that point on the tire
a tire in a yaw. moves towards the rear of the contact patch before lifting
away form the ground at Position 2. All of the parameters
VT − ωr that appear in Equation (5), V, α, ω, and r are visible in
S= (4) Figure 5.
VT
The time it takes to deposit one striation mark, t, will now
V cos α − ωr be explored. Equation (5) can be rewritten as a function
S= (5) of this time, t, to yield Equation (6). Time is equal in all
V cos α terms of Equation (6) and cancels out. Rearranging
results in Equation (7).
Now consider Figure 5. While the shoulder tread block of
a yawing tire deposits the striation mark indicated by the
bold line, it moves from Position 1 to Position 2 up the
page. The two images on the right depict a side view of

4
 d βr sin θ
cos α − S = 1− (14)
S= t t (6) (cos θ sin α + sin θ cos α )cos α
d
cos α
t sin θ
S = 1− (15)
sin(θ + α ) cos α
βr
S = 1− (7)
d cos α sin θ
S = 1− (16)
Two different triangles can be used to equate the tire sinψ cos α
mark width, as shown in Equation (8). The contact patch
length can then be determined using Equation (9). The sin(ψ − α ) (17)
images on the left of Figure 5 reveal that the contact S = 1−
sinψ cos α
patch length is equal to βr. It should be noted that this
contact patch length calculated is theoretical because
only the shoulder blocks are assumed to be leaving  sinψ cos α − cosψ sin α 
marks on the road. In actuality, the bottom tread of the S = 1 −   (18)
 sin ψ cos α 
tire also leaves marks and the contact patch length is not
so readily determined. For the modeling that follows, the
triangles would be of different proportions if the actual cosψ sin α
contact patch length were utilized. As will be shown, all
S = 1 −1 + (19)
sinψ cos α
the triangle side lengths fall out of the analysis leaving
the trigonometric relationships which are unaffected by
the shoulder tread block assumption. tan α
S% = ×100 (20)
tan(θ + α )
CPL sin α = w = LS sin θ (8)
Equation (20) is multiplied by 100 giving slip percentage,
LS sin θ S%. Equation (20) is mathematically equivalent to the
βr = CPL = (9)
classic equation of slip during yaw, Equation (5).
sin α
Equation (20) demonstrates that the slip percentage of a
tire during the depositing of a striation mark can be
As is seen in Figure 5, d is equal to the summation of d1
calculated if the angle of that striation and the slip angle
and d2 which can be determined geometrically, as
of the tire are known.
follows:
Next, consider the effects of braking on tire mark/tire
d1 = LS cos θ (10) geometry. Figure 6 depicts three scenarios; a free rolling
tire on the left, partial braking in the center and full wheel
LS sin θ cos α (11)
lockup on the right. The vertical line in each scenario,
d 2 = CPL cos α = which is aligned to the tires velocity direction, has a
sin α dimensionless length of 1. From the bottom end of this
vertical line, two lines are drawn, one along the heading
LS sin θ cos α direction of the tire, an angle α from the velocity direction,
d = d1 + d 2 = LS cos θ + (12)
sin α and the other along the direction of the striation, θ from
the tire velocity. Two lines are then drawn from the top
Equation (9) and Equation (12) can then be substituted end of the vertical line that intersect perpendicularly with
into Equation (7) yielding Equation (13). the tire heading and striation direction. In all three
scenarios, the tire has the same slip angle. The striation
directions, however, are different in each case, ranging
 LS sin θ 
  from perpendicular to the tire in the no braking case, to in
S = 1−  sin α  (13) line with the velocity direction in the full braking case.
 LS sin θ cos α 
 LS cos θ +  cos α
 sin α  The ratio of the lengths of dactual and dfreeroll, which are
labeled in Figure 6, also define the amount of slip of the
Equation (13) can then be simplified through the following tire. As is shown in the partial braking case, the vector
series of equations to obtain Equation (20). dactual is parallel to dfreeroll and becomes shorter in length
as the angle θ decreases, which occurs as braking
increases. Because dactual is parallel to dfreeroll, an angle of
Ψ separates it from the striation direction.
5
 Figure 6 – Braking triangles.

Equation (21) defines the amount of slip in terms of the sin(ψ − α )

S = 1− (26)
geometry presented in Figure 6.
sinψ cos α
d actual (21)
Slip = S = 1 − sinψ cos α − cosψ sin α (27)
d free _ roll S = 1−
sinψ cos α
For the free rolling case, dactual and dfreeroll are of equal
length resulting in a slip value of zero. The full lockup sin α cosψ (28)
S=
case, on the other hand, dactual is zero resulting in a slip sinψ cos α
value of 1, or 100%. Equation (21) and Figure 6, describe
mathematically the idea that striation marks are
perpendicular to the tire in the absence of braking and tan α (29)
S=
parallel to the velocity direction in the presence of full tanψ
wheel lock up. Equation (21) also offers insight into the
cases where partial braking is occurring.
tan α (30)
S% = ×100
The triangles in the partial braking case in Figure 6 can tan(θ + α )
be used to find expressions for dactual and dfreeroll, as
follows: STRIATION STUDY I: TIRE MARKS DEPOSITED
FOLLOWING AN IMPACT
sin θ = d actual sinψ (22)
The simplest case examined during this research is that
sin θ of a rear tire with no braking. As discussed in earlier
d actual = (23) sections, when no braking is present during a yaw,
sinψ striations are deposited in a direction perpendicular to the
heading of the tire. For most vehicles, which are steered
d freeroll = cos α (24) by only the front tires, the heading direction of the rear
tires is the same as the heading direction of the vehicle.
Equations (23) and (24) can be substituted into Equation Thus, in theory, the slip angle of a rear tire with no
(21), yielding Equation (25). Equation (25) can then be braking can be determined by simply aligning the tire in a
simplified to Equation (30), which again is equivalent to direction perpendicular to a striation mark. Further,
the classic equation for slip. striations deposited by a rear tire can allow the
reconstruction of an entire yaw sequence from a single
sin θ (25) tire mark with relative ease, so long as no braking was
S = 1− occurring during the yaw.
sinψ cos α
One scenario where there is often no braking, at least no
braking input by the driver, is in a yaw induced by an
rd
impact. On July 23 , 2008, the authors conducted two
6
Pursuit Intervention Technique tests, or PIT tests, at the rear quarter panel of the suspect’s vehicle. The officer
Denver Police Academy. The PIT maneuver is a method then steers their vehicle into the rear of the suspect
used by police to end a pursuit by spinning out a vehicle, causing the suspect to spin out. Figure 7 depicts
suspect’s vehicle. The pursuing officer positions their images from one of the PIT tests conducted by the
vehicle to one side of a suspect’s vehicle such that the authors.
front quarter panel of the police vehicle aligns with the

Figure 7 - PIT Test 1

Both vehicles used in the PIT tests were Ford Crown tire marks deposited by the Crown Victoria. Based on
Victorias that had been modified by the Denver Police Figure 10, we can conclude that the post-impact yaw
Department specifically for PIT training. The target motion of a vehicle can be reconstructed using a single
vehicle in the test, the PITed vehicle, had a roll cage striated tire mark, so long as sufficient evidence is
mounted to its interior. It also had a steel frame over its available to allow the analyst to identify which tire
rear bumper and rear fenders to prevent damage during deposited the single tiremark. This also confirms that in
PIT maneuvers. Likewise, the bullet vehicle, or PITting the absence of braking, striations are deposited in a
vehicle, had a steel frame covering its front bumper and direction perpendicular to the tire.
front fenders to prevent damage.

After the PIT maneuver was initiated, the driver of the

PITed vehicle released his hands from the steering wheel
and removed his foot from the pedals. The PITed vehicle
yawed counterclockwise in the first test and clockwise in
the second. After each test, the tire mark evidence and
rest position of the vehicle were surveyed and
photographed. Figure 8 depicts the tire marks deposited
during the second test. Tape was used to identify
individual striation marks which were included in the
survey. The vehicle that deposited the tire marks in
Figure 8 was travelling out of the page, or from the top of
the photograph to the bottom.

Figure 9 depicts several positions of the PITed Crown

Victoria that have been aligned to the surveyed striations
in the rear left tire mark. At each position, the point on the
outside edge of the rear left tire directly below the axle
was aligned to the midpoint of the striation. The vehicle Figure 8 - Striations deposited during the PIT Test 2
was then rotated in yaw until the heading of the tire was
perpendicular to the striation. Figure 10 depicts the same
vehicle positions and also includes the other 3 tire marks

7
 nd
Figure 9 - Motion from 2 PIT Test. Vehicle positions determined by aligning the left rear tire perpendicular to striations
within the left rear tire mark. The remaining 3 tire marks have been hidden.

Figure 10 – Identical to Figure 9, with the exception that all the tire marks are shown.

STRIATION STUDY II: TIRE MARKS DEPOSITED

FOLLOWING A LOSS-OF-CONTROL

In order to assess ability of striations to offer insight into

braking and steering, the authors conducted yaw testing
with a 2008 Chevrolet Malibu on August 18, 2008. The
Malibu was outfitted with Hankook Winter Ipike tires, of
size 225/60R16. The vehicle was also instrumented with
data acquisition equipment from Racelogic. Specifically
the VBOX IISX + Slip, Pitch and Roll Angle recorded the
vehicle’s angular and translational position and the
Vehicle CAN Interface recorded wheel speeds and
steering position from the Malibu’s internal computer. All
data was recorded at 20 Hz. Figure 11 depicts the Malibu
test vehicle with the three VBOX GPS antenna attached
magnetically to its roof. A professional driver was utilized Figure 11 – Test Chevrolet Malibu
for this testing.

8
During our testing with the Malibu, several attempts were angle. The plot in Figure 12 depicts the tire angle versus
made to induce a yaw using only steering inputs. the steering angle recorded by the VBOX for the front left
Although tracking of the rear tires outside the front tires and front right tires across the full range of steering of the
did occur during these attempts, yaw was insufficient to Malibu. As can be seen in Figure 12, right and left tires
deposit striations that were distinguishable enough for are different, largely due to the Ackerman geometry.
this study. In order to induce yaw rotation of the Figure 13 depicts three images from one of the Malibu
magnitude needed for this study, it was necessary to yaw tests.
apply the parking brake temporarily to initiate the yaw.
After the yaw was initiated, the parking brake was
released and striations were deposited using
combinations of steering and braking.

The position of the vehicle at its starting and rest

positions were surveyed. The tire marks and test surface
were also surveyed and photographed and a test scene
diagram was created. Individual striation marks were
surveyed and documented photographically.

Following the test, the orientation of the front left tire

associated with various steering wheel angles was Figure 12 – VBOX recorded steering angles vs. tire
surveyed across the full range of steering. During this angles through the full range of steering.
process, the steering angles where recorded via the
VBOX CAN Interface. This allowed for the authors to
correlate the VBOX recorded steering data to the tire

Figure 13 - Yaw Test 2 of Chevrolet Malibu.

As mentioned earlier, shoulder tread block sizes and marks. Given that the yaw angle of a rear tire is the same
spacing are variable on actual tires. The characteristics as the yaw angle of the vehicle, the slip percentage can
of tire marks, as one would expect, are heavily influenced be easily calculated using Equation (30). This equation
by the characteristics of the tires depositing the marks. only requires that the slip angle of the tire and the
The tires used in these tests were chosen specifically for striation angle are known. This implies that if a vehicle
their large block shoulder tread pattern, which is similar to can be aligned with multiple tire marks, the amount of
tires that are regularly seen on SUV’s and light trucks braking of a rear tire can be determined with a single
and, which produce tire mark striations that are easier to striation. For front tires, both braking and steering are of
identify. Each tire had 58 shoulder tread blocks around its interest so the spacing between the striations must be
circumference, which has implications in the analysis that considered.
follows. Figure 14 depicts the shoulder tread blocks as
well as the shoulder wear on the front left tire as result of The results from analysis of the striation marks in both
the testing. tests were compared to analysis of the VBOX recorded
data. To achieve this comparison, it was necessary to
The purpose of these tests was to determine the link the VBOX data to the tire mark evidence on the
feasibility of using tire mark striations to predict braking roadway. This link was made possible through animation
and steering. For this vehicle, as is typically the case, of the VBOX recorded angular and translational position
only the front tires are steered. Thus, different analysis of the vehicle. This data was animated over the top of the
techniques can be used depending on whether a rear or surveyed tire mark evidence which effectively linked the
front tire is of interest. In most occasions, as was the VBOX data and tire mark evidence in time. Once linked,
case in these tests, multiple tire marks are deposited VBOX data could be examined at each vehicle location of
when a vehicle yaws and the vehicle’s yaw angle can be interest.
determined by alignment of the vehicle with the tire
9
 accelerated to a speed of approximately 48 mph. The
parking brake was then applied and the driver made a
steering input to the right initiating a clockwise yaw. The
parking brake was then released and the driver applied
the service brakes and counter-steered aggressively to
the left. During the event, the vehicle yawed a total of
approximately 90 degrees (see Figure 15). The driver
stated that he braked significantly but that the ABS did
not engage during the test. Upon inspection of the wheel
speed data it was found that the ABS indeed did not
engage during this test.

Seven positions along the path of the vehicle were

analyzed. Six striation marks in the left rear tire mark
were first examined with braking as the focus. A portion
of the front left tire mark was then examined to determine
steering and braking from the tire mark evidence. The
Figure 14 - Front left tire from Malibu.
positions of the Malibu analyzed are depicted in Figure
15. Position 4 is the approximate location of analysis of
TEST II
the front left tire. Three of the six positions of the left rear
tire mark that were analyzed are indicated with tape in
The results of the testing are more easily described
Figure 16. In Figure 16, the Malibu was travelling into the
beginning with the second test. In Test II, the Malibu was
page.

Figure 15 – Tire mark evidence, vehicle positions and rest position from Test 2.

10
 (5), shown in slightly different form below, to yield the
VBOX slip percentage of the tire at the 6 locations.

VTranslational cos α − VWheelSpeed

S %VBOX = × 100 (31)
VTranslational cos α

The slip percentage was then calculated using the

striation marks on the roadway with Equation (30),
restated below, for comparison with the results from the
analysis of the VBOX slip percentage.

tan α
S % STRIATION = × 100 (32)
tan(θ + α )

Again, α is the tire slip angle and θ is the striation angle in

Equation (32). The slip angle at each position was
previously calculated. The striation angle, which is the
angular difference between a striation and the tire
velocity direction at each respective location, was
measured. The results of the slip percentage analysis at
the 6 positions on the rear left tire mark in Test 2 are
shown in Figure 17. The dashed line and solid line in
Figure 17 connect the results from analysis of VBOX data
with Equation (31) and results from the striation analysis
with Equation (32), respectively.

In Test 2 the parking brake was engaged in combination

with a right steering input. Once a clockwise yaw was
initiated, the parking brake was released and the driver
Figure 16 - Left rear tire mark from Test 2. input aggressive counter-steering to the left and applied
the service brakes. The sequence of the driver’s braking
Prior to performing calculations using the striation data, actions is reflected in the plot of Figure 17. During the
the VBOX collected data was analyzed for comparison first position, the driver was in the process of releasing
purposes. To compute the slip percentage of the rear left the parking brake, which resulted in approximately 60%
tire using the VBOX data at each position, the slip angle, slip. The next three positions have slip near 0 % and
translational speed and rotational speed of the tire were represent the time between the release of the parking
required. The slip angle of the tire at each position was brake and application of the service brakes. The slip
determined based on alignment of the vehicle on the tire percentage then increases to a value of over 10% by the
th
marks. Since the rear tires cannot be steered, the 6 position as result of the service brake application.
heading direction of the rear tires are approximately the
same as the heading direction of the vehicle. The slip
angle of the tire, which is the angular difference between
the tire’s heading and tire’s velocity direction, was
measured. The translational speed of the rear left tire
was calculated considering both the translational velocity
recorded by the VBOX at the antenna location and the
yaw velocity of the vehicle.

The rotational speed of the wheel output by the VBOX

was given in feet per second, rather than a rotational
velocity unit, due to a conversion performed by the
vehicle computer, specifically, multiplication by a tire
radius, r. The tires used during our testing were of slightly
different radius than that used to convert the data by the
vehicle computer. Therefore, the wheel speed data was Figure 17 – Results from slip percentage analysis of the
scaled such that during straight line motion, when slip of rear left tire in Test II.
the tires was zero, the translational velocity and pre-
converted rotational velocity of the tire were equal. These
values could then be inserted into Equation (5), shown in
11
The portion of the front left tire mark to be analyzed is  SDTOTAL sin θ 
shown in Figure 18, in which the vehicle was moving from α = sin −1   − θ (33)
left to right, across the page. In Figure 18, 58 striations  CT 
have been identified and marked with chalk. As
mentioned earlier, the tires used in this test had 58 tread Equation (33) was analyzed for the 58 striations which
blocks around their circumference. Therefore, over the appear in Figure 18 using Monte Carlo Analysis. Monte
course of depositing the 58 striations in Figure 18, the tire Carlo analysis was used to monitor the potential
rotated one revolution. The average slip angle and measurement errors in the striation direction and
average braking over the course of one tire revolution will spacing. The overall length of the 58 striations, SDtotal,
be studied here. was varied by ±1 inch uniformly. The average striation
st th
angle, θ, between the 1 and 58 striations was used,
Consideration of one tire revolution offered two benefits and varied uniformly ±1 degree. The circumference of the
to the analysis. First, the tread blocks around the tire, CT, was measured at the tire shoulder. Since the
circumference are variable both in their size and in the steering angle of the tire is the difference between the
spacing between them. To analyze the slip angle using yaw angle of the vehicle, and the slip angle of the tire, α,
less than 58 striations would have required knowing the yaw position of the vehicle was needed. The yaw
which individual tread blocks on the tire deposited each
position at the midpoint of the length SDTOTAL was
mark. If an entire revolution is considered however,
determined by alignment with the tire marks and varied
Equation 2 can be rewritten as follows, where CT is the
uniformly by ±1 degree.
circumference of the tire, and SDTOTAL is the distance on
the ground over which striations of number equal to the
The steering angle was recorded by the VBOX CAN
total number of tread blocks on the tire are deposited, as
interface from the Malibu’s onboard computer. The VBOX
depicted in Figure 18. Analysis of one revolution also
slip percentage was calculated in similar manner to that
made the distances to be measured much larger, which
of the rear tire, with the exception that slip angle was
reduced the sensitivity of the analysis to errors in
determined using both the yaw position of the vehicle as
measurements.
aligned with the tire marks and the VBOX recorded
steering angle.

Figure 18: Front left tire mark from Test 2.

12
The results of the Monte Carlo analysis are displayed in counter-steering to the left, consistent with the actions of
the three distribution charts of Figure 19. In each plot, the the driver. The VBOX recorded steering angle of 7.8
actual value, as calculated with the VBOX data, is degrees at the tire, or 124 degrees at the steering wheel,
indicated with a vertical line, and numeric value. The first was encompassed in the range of values calculated from
plot displays the range of values for the angle of the tire. the striation marks.
The second plot displays the steering wheel angle and
was calculated by multiplying the angle of the tire by the The VBOX calculated slip percentage of approximately
steering ratio, which is approximately 16.4 for this 27.5 did not fall within the range of computed values,
2
particular vehicle. The third plot displays braking in the which ranged from 19.5 to 22.7. However, the results of
form of slip percentage. both the VBOX recorded and striation calculated slip
percentages indicate significant braking at this analyzed
position.

TEST I

In Test I, the Malibu was again accelerated to a speed of

approximately 48 mph. The parking brake was applied
and the driver made a steering input to the left initiating a
counter-clockwise yaw. The parking brake was then
released and the vehicle yawed approximately 180
degrees. Once the parking brake was released, no
additional braking occurred and no attempts at counter-
steering were made by the driver. The authors analyzed
one section of the front right tire mark, 58 striations in
length, shown in Figure 20. The vehicle was travelling
into the page in Figure 20.

Figure 19: Results of the analysis of Test 2.

As can be seen in the first two plots, a large range of

steering angles resulted from the variance of the inputs.
However, all computed values indicated significant
Figure 20: Front right tire mark from Test I.
2
Manufacturer specifications.
13
Similar to Test II, the striations and the data recorded by arcsine results in an error. In the context of our analysis,
the VBOX were analyzed and compared. According to the same range was used for the input variables in both
the VBOX data, at the time this mark was deposited, the tests, but the output ranges in Test I were much larger.
steer angle at the front right tire was negligible,
approximately ½ degree to the left. The driver input no At first glance, this sensitivity in the no braking case
braking after the parking brake was released so the slip appears to present a problem to striation analysis.
percentage throughout the yaw was zero. However, the fact that, at least in these two cases, the
model appears to return satisfactory results in the case of
The striation marks were analyzed in the same manner braking and overly sensitive results in the case of no
as Test II. All inputs were varied by the same amount. braking can actually be of advantage. In the analysis of
Unlike Test II, the results of the analysis of the striations Test I, it was found that a value SDTOTAL less than 1 inch
in Test I were not in agreement with the analysis of the greater than the measured value returned an error. By
VBOX results. The striations indicated aggressive contrast, in the analysis of striations in Test II, the length
rightward steering, 40 to 240 degrees at the steering SDTOTAL could be increased by over 10 inches before
wheel when in reality the steering was negligible at this errors were returned. The fact that the results in Test I
time. resulted in errors with minor changes to SDTOTAL gave
good indication the tire was near zero percent slip. The
The striations also indicated slip percentages between 0 sensitivity of these equations will be discussed in greater
and 13 percent when in reality no braking was occurring. detail later.
The root cause of this disagreement was found to be the
sensitivity of Equation (33) in the no braking case, when For now, recall that in the no braking case striations are
the term identified as Equation (34) approaches 1. deposited perpendicular to the tire. If the proximity of the
calculated slip angle to an error output was used to
 SDTOTAL sin θ  assume that no braking occurred at the time, the tire
  (34) could simply be aligned perpendicular to the striation
 C T  mark to determine the slip angle, α. The authors used
st
this approach and aligned the tire perpendicular to the 1
th
In calculating the slip angle with Equation (33), the and 58 striations yielding tire angles of 2.2 degrees and
arcsine of the term above is evaluated. At values near 1, 0.5 degrees respectively. Again, the steering angle of the
small changes in the term above result in large changes tire as recorded was 0.5 degrees. Figure 21 depicts the
in the arcsine of the term. When this term equals 1, the position of the vehicle at these two positions, as well as
arcsine is equal to 90 degrees. At values over 1, the the rest position.

Figure 21 – Tire mark evidence, vehicle positions and rest position from Test 1.

14
DISCUSSION was occurring at that tire. In the no braking case, the tire
could simply be aligned perpendicular the striations,
It was found that information exists within yaw mark yielding the slip angle, and subsequently, the steering
striations that can offer insight into the actions of the angle. The results of Test 1 revealed the sensitivity of
driver at the time the yaw marks were being deposited. It Equation (33) in the no braking case. The sensitivity of
was also confirmed that it is possible to reconstruct post- Equation (33) is also dependant on the striation angle, θ,
impact yaw motion with a single striated rear tire mark. with larger striation angles increasing sensitivity. In
Although challenges arose from sensitivities in the comparing the results of Test I and II, the overly sensitive
measurements and in the equations, it was still possible slip angles calculated in Test I allowed for a no braking
to draw generalized conclusions from the striations, such determination to be made. Caution should be observed,
as whether the driver was steering significantly in one however, in establishing no braking in this way. A full
direction or if the brakes were applied. sensitivity study of these equations is needed before the
results of these two tests can be generalized over a
It is possible that refinements in measurement and complete range of striation angles. This sensitivity study
analysis techniques could improve the results of striation is currently being conducted by the authors.
analysis. In these tests, only the leading side tires
deposited identifiable striations. It is often the case The ABS did not engage in either of the tests. Since the
though, that leading and trailing side tires both deposit rotational speed of the tire directly contributes to the
identifiable striated marks. If one were to analyze both striations that are deposited, it would be interesting to
the right front and the left front tire mark striations, and include a case where the ABS engaged in future testing.
could reach some commonality, greater confidence in Similarly, it would be worthwhile to include vehicles
calculated steering angles could potentially be achieved. equipped with ESC in a future study.
These authors will be experimenting with test conditions
in the attempt to deposit distinguishable striated tire As described earlier, the equations presented in this
marks at all four tires. paper assume a rigid tire. In reality, the tire does deform
extensively in some high slip angle maneuvers that
It was found that the most significant limitation in deposit striated tire marks. In the two tests analyzed in
determining steering angles was the fact that it is not the this study, errors due to sensitivities in the equations
angle of the tires, but angle of steering wheel that is of make it unclear how tire deformation affected the results.
interest. The steering ratio of the Malibu, or ratio between It is possible that modifying the model to include
the angle of the steering wheel and angle of the tires, deformation would improve the results. It is also possible,
was approximately 16.4:1. This being the case, any however, that uncertainties that arise in modeling the
errors that existed in determining the angle of the tire deformation of the tire could make the model less
were multiplied by 16.4 when examining the steering accurate. In any case, there were some findings from this
wheel. research that may be useful considerations for tire
modeling theory. For one, the striations do indicate the
In the case of braking, it was slip percentage that was direction of the friction force. This coupled with the ability
calculated in this study. In the context of accident to determine the approximate slip angle from the tire
reconstruction though, it is the rate of deceleration of the marks could be useful information for modeling tire
vehicle that is of primary interest. The manner in which behavior.
deceleration and slip percentage are related is non-linear
and is surface, tire and slip angle dependant. The authors Consider again Figures 1 and 3, combined below in
do not intend to take up a full treatment of the topic here. Figure 22. In the tire mark on the left, a clear distinction
For the purpose of this paper, suffice it to say that, in between light and dark striations is visible on the tire
Test I, the test without braking, the vehicle decelerated at mark. The tire marks deposited in the testing performed
an average rate of approximately .46 g’s. By comparison, in this study did not display a similar light/dark distinction
the driver reported braking aggressively in Test II, and likely due to tire and test condition differences. In tire
the vehicle decelerated at an average deceleration of .68 marks that are deposited prior to actual accidents, this
g’s. In Test II, both the VBOX and the striations indicated light/dark distinction is often present. The graphic on the
that application of the brakes resulted in 10 to 20 percent right describes how one could potentially align a tire on a
slip at the rear left tire and approximately 28 percent slip tire mark similar to the one on the left after determining
at the front left tire. The fact that braking during a yaw the slip angle using the methods presented in this paper.
produces different amount of slip at each tire presents
another interesting topic for future work.

The results of the analysis of Test I deserve more

attention. Recall that the slip angle equation, Equation
(33), was extremely sensitive near the case of no
braking. The authors proposed using this sensitivity to
their advantage as a way to establish that no braking was
20
 Kevin Wells, Toby Terpstra, Will Bortles, Mike Farr, and
Mike Walker for their assistance in the vehicle testing.
We would also like to thank Jim Anderson, Mike Farr and
Mike Walker for their excellent driving. Finally, the
authors would like to thank the Denver Police Department
for the use of their testing grounds.

REFERENCES

1. Daily, John, Shigemura, Nate, Daily, Jeremy, “Linking

Road Evidence to Vehicle Performance By Using A
Novel Paint Ball Marking System”,
http://www.jhscientific.com/DRMS/index.html.

2. NHTSA, “An Experimental Evaluation of 26 Light

Vehicles Using Test Maneuvers That May Induce
On-Road, Untripped Rollover and a Discussion of
NHTSA’s Refined Test Procedures: Phases 6 and 7
of NHTSA’s Light Vehicle Rollover Research
Program.” October 2003

3. Gillespie, Thomas D., “Fundamentals of Vehicle

Dynamics.” SAE, 1992.
Figure 22
4. Reveley, Mary S., Brown, Douglas R., Guenther,
Alignment of a tire on a tire mark with a light/dark Dennis A., “A Comparison Study of Skid and
distinction in this way inherently reveals the actual Yawmarks.” SAE, 890635.
contact patch length of the tire. The contact patch length
could certainly be of use in modeling the deflection of a 5. Baker, Kenneth S., “Traffic Collisions Investigation.”
tire since contact patch length is directly related to Northwestern University Center for Public Safety,
deformation. Contact patch length is also directly related 2001.
to tire inflation pressure. The authors are currently
investigating the effect tire under inflation has on the 6. Daily, John, Shigemura, Nathan, Daily, Jeremy.
appearance of deposited striations. “Fundementals of Traffic Crash Reconstruction.”
IPTM, 2006.
In all of the analysis presented above, it was necessary
to align the vehicle over its tire marks. Due to two factors, 7. Milliken, William F., Milliken, Douglas L. “Race Car
tire deformation and steering of the front tires, alignment Vehicle Dynamics.” SAE, 1995.
of the tires to the tire marks required some judgment on
the part of the analyst. Tire alignment over tire marks is
an area in need of more research and is another focus of CONTACT
ongoing work.
Gray Beauchamp
ACKNOWLEDGMENTS Kineticorp, LLC
6070 Greenwood Plaza Blvd., Suite 200
The Authors would like to thank Dr. Ronald Rorrer from Greenwood Village, Colorado 80111
the University of Colorado, Dr. Patrick Fitzhorn from (303) 733-1888
Colorado State University and William Neale from [email protected]
Kineticorp for their helpful comments. Also, thank you to www.kineticorp.com

21
APPENDIX A: NOMENCLATURE

a Tire slip angle

Ψ Angle between tire heading and striation mark
θ Angle between tire mark direction or wheel hub velocity
SD Distance from leading edge of one striation to the leading edge of an adjacent
striation, measured along tire mark direction, or wheel hub velocity direction
TD Distance from leading edge of one striation to the leading edge of an adjacent
striation, measured perpendicular to striation
V Translational velocity of wheel hub

β Angle rotated by tire during one point on the tire’s contact with
the ground
ω Rotational velocity of tire
CPL Contact patch length
d Summation of d1 and d2
d1 Length of striation mark measured along the tire mark direction or
wheel hub velocity direction
d2 Contact patch length measured along the tire mark direction or
wheel hub velocity direction
LS Absolute length of striation
r Tire radius
VT Translational velocity of wheel hub along tire heading
W Tire mark width

CT Tire Circumference
S Braking slip
th
SDTOTAL Distance from leading edge of one striation to the leading edge of the x later striation, measured
along tire mark direction, or wheel hub velocity direction
S% Braking slip percentage
S%STRIATION Braking slip percentage calculated from striations
S%VBOX Braking slip percentage calculated from VBOX recorded data
VTranslational Translational velocity of wheel hub, calculated from VBOX recorded data
VWheelSpeed Converted wheel speed from VBOX, given in mph

22