2

Kinematics Fundamentals

2.1 Introduction
This chapter discusses the kinematics of motion, that is, looking at the nature of the motion
without examining the forces that cause the motion. We will focus on the two- and three-
dimensional kinematics of particles, as well as planar kinematics of rigid bodies. Three-
dimensional kinematics of rigid bodies will be discussed in Chapter 9.
We begin with coordinate systems and the kinematics of particles. The motion of par-
ticles is purely translational. Then, rotating reference frames, rotation parameters, angular
velocity, and angular acceleration are discussed. Relative velocity and acceleration equations
are developed. Instant centers are introduced, as they are crucial to the analysis of vehicles
and mechanisms.
Kinematic analysis serves two purposes: First and foremost, it is a precursor to kinetic
analysis, a topic that will be discussed in the next chapter. We cannot analyze the kinetics
of a system without first studying its kinematics. In addition, kinematic analysis by itself
is a valuable tool and is widely used in the design of mechanisms and vehicle suspensions,
as well as in motion planning. Chapter 3 discusses applications of kinematics.

2.2 Position, Velocity, and Acceleration

When studying the kinematics of a particle, that is, the translational motion of a point, we
need to describe its position, velocity, and acceleration. The description must be made with
respect to a reference point or origin using a coordinate system. A coordinate system or
coordinate frame is characterized by a set of coordinate axes, the positive directions of these
axes, and unit vectors along these axes. Several types of coordinate systems exist. We select
the coordinate system (or frame) that will make the analysis easier and more meaningful.
Selecting a coordinate frame is, in essence, selecting the set of motion variables.
Figure 2.1 depicts a particle that is moving. The position is measured from a reference
point that is fixed, and the position vector is denoted by r (t). Consider next the position
of the same particle at a time increment ∆t later, at time t + ∆t. The position vector at
t + ∆t is r (t + ∆t). The velocity v (t) and acceleration a (t) of the particle are defined as

dr ∆r r (t + ∆t) − r (t)
v (t) = = lim = lim
dt ∆t→0 ∆t ∆t→0 ∆t

dv v (t + ∆t) − v (t)
a (t) = = lim (2.1)
dt ∆t→0 ∆t
The rate of change of acceleration is of interest in several applications. Examples include

25
26 Applied Dynamics

FIGURE 2.1
A particle and its path.

vehicle dynamics and human motion analysis. The commonly used terms for the rate of
change of acceleration are shock are jerk. The occupants of a vehicle get shaken and thus
experience discomfort if the acceleration profile undergoes a sudden change, such as when
accelerating or braking rapidly and when taking sharp turns. A vehicle and its components
wear out sooner when they are repeatedly subjected to sudden accelerations.

2.3 Reference Frames: Single Rotation in a Plane

Coordinate systems are used in kinematics to observe motion. We decide on which coordi-
nate system to use by considering the nature of the motion. This section develops relation-
ships between different coordinate systems and transformations from one coordinate system
to another. Only right-handed coordinate systems are considered here. When we point our
right hand towards the positive direction of one of the axes (say, x, with unit vector i) and
rotate our fingers towards the positive direction of the second axis (say, y, with unit vector
j), the thumb points in the positive direction of the third axis z, with unit vector k = i × j.

Y
y
sin

J

cos
x
j
i sin



X
cos
I

FIGURE 2.2
The XY and xy coordinate systems.

Consider plane motion and a planar coordinate system XY , as shown in Figure 2.2. The
Kinematics Fundamentals 27

unit vectors along the X and Y directions are I and J, respectively. Also shown in the same
figure is a coordinate system xy (with unit vectors i and j) that is obtained by rotating
the XY axes by an angle θ in the counterclockwise direction. The rotation is about the
Z axis (not shown here) perpendicular to the plane. The Z and z axes are the same. The
relationship between the unit vectors of the two coordinate systems is
i = cos θI + sin θJ j = − sin θI + cos θJ

I = cos θi − sin θj J = sin θi + cos θj (2.2)

The unit vector perpendicular to the plane of motion is common to both coordinate systems,
so that k = K. The above relationships can be expressed in matrix form as
         
i cos θ sin θ I I cos θ − sin θ i
= = (2.3)
j − sin θ cos θ J J sin θ cos θ j
Define the rotation matrix [R] as
 
cos θ sin θ
[R] = (2.4)
− sin θ cos θ
The matrix [R] is unitary; its determinant is equal to 1 and its inverse is equal to its trans-
pose, [R]−1 = [R]T . The two relationships in Equation (2.3) are inverse transformations.
Next, consider a point P on the plane (Figure 2.3) and express the coordinates of point
P as (XP , YP ) in the XY coordinates and (xP , yP ) in the xy frame, as shown in Figure 2.3.
A set of independent coordinates that describe the orientation of a system completely are
called generalized coordinates. The vector rP , which denotes the position of point P , can
then be written in terms of the two coordinate systems as
rP = XP I + YP J = xP i + yP j (2.5)

Y
y
YP P

yP

rP

J x
j
i xP

X
I XP

FIGURE 2.3
Representation of vector rP in XY and xy coordinates.

It is of interest to explore the relationship between the components of the two descrip-
tions of rP . To this end, introduction of Equation (2.3) to the above equation results in
rP = XP I + YP J = XP (cos θi − sin θj) + YP (sin θi + cos θj)
28 Applied Dynamics

= (XP cos θ + YP sin θ) i + (YP cos θ − XP sin θ) j = xP i + yP j (2.6)

from which we conclude that
xP = XP cos θ + YP sin θ yP = YP cos θ − XP sin θ (2.7)
At this stage, we introduce the column vector representations of the unit vectors and of
the location of point P in the two reference frames:
   
xP XP
{xyz rP } =
XY Z
rP = (2.8)
yP YP
and express the transformation between the two column vectors as
   
xP XP
{xyz rP } = [R] XY Z rP

= [R] or (2.9)
yP YP
The inverse relationship between the two vectors is
rP = [R]−1 {xyz rP }
XY Z XY Z
or rP = [R]T {xyz rP } (2.10)
The above relationships can be generalized to three dimensions by redefining the column
vector representations and the rotation matrix as
     
xP XP cos θ sin θ 0
{xyz rP } =  yP  XY Z

rP =  YP  [R] =  − sin θ cos θ 0  (2.11)
zP ZP 0 0 1
We see that the same relationship that governs the position vectors is also valid for
the unit vectors. Indeed, expressing the unit vectors in column vector format [i j k]T and
[I J K]T , we can write
       
i I I i
 j  = [R]  J   J  = [R]T  j  (2.12)
k K K k

Example 2.1
The XY Z coordinate system is rotated about the Z axis by 30◦ clockwise to obtain the
xyz coordinates. Consider the vector r = 3i − 4j and express it in terms of the XY Z frame.
Using column vector notation, r is
 
xyz 3
{ r} = [a]
−4
The transformation angle is θ = −30◦ , so that the matrix between the two coordinate
systems is    
cos θ sin θ 0.8660 −0.5000
[R] = = [b]
− sin θ cos θ 0.5000 0.8660
The second part of Equation (2.10) can be used to express r in the XY Z frame, with
the result
    
XY Z 0.8660 −0.5000 3 0.5981
r = [R]T {xyz r} = = [c]
0.5000 0.8660 −4 −4.9641

The results can be checked by noting that the magnitude of {xyz r} and XY Z r must
be the same. Therefore,
XY Z T XY Z
{xyz r}T {xyz r} = 32 + 42 = 25 r r = 0.59812 + 4.96412 = 25.00001
[d]
where the difference is due to roundoff error.
Kinematics Fundamentals 29

2.4 Column Vector Representation

This section discusses two notations to represent vectors. Consider a coordinate system with
unit vectors e1 , e2 , and e3 , which form a mutually orthogonal set. Also consider two vectors
r and q defined as

r = r1 e1 + r2 e2 + r3 e3 q = q1 e1 + q2 e2 + q3 e3 (2.13)

Vectors described this way are referred to as geometric vectors or spatial vectors. The dot
and cross products of these vectors yield

r · q = r1 q1 + r2 q2 + r3 q3

r × q = (r2 q3 − r3 q2 ) e1 + (r3 q1 − r1 q3 ) e2 + (r1 q2 − r2 q1 ) e3 (2.14)

The previous section demonstrated that the vectors r and q can be expressed in column
vector format as
   
r1 q1
{r} =  r2  {q} =  q2  (2.15)
r3 q3

The column vectors are also referred to as algebraic vectors. Using this description, we
can express the dot product of two geometric vectors in column vector format as

r · q =⇒ {r}T {q} (2.16)

The skew-symmetric matrix [r̃] associated with the column vector {r} is a compact way of
expressing a cross product. Define it as
 
0 −r3 r2
[r̃] =  r3 0 −r1  (2.17)
−r2 r1 0

so that
 
r2 q3 − r3 q2
r × q =⇒ [r̃] {q} =  r3 q1 − r1 q3  (2.18)
r1 q2 − r2 q1

Note that because r × q = −q × r, the relationship [r̃] {q} = − [q̃] {r} also holds.
In kinematics we frequently encounter the vector product r × (r × q). The expression is
commonly shortened to r × r × q, with the understanding that the cross product between
r and q is performed first. Using the notation introduced above,

r × (r × q) =⇒ [r̃] [r̃] {q} (2.19)

The matrix multiplications in [r̃] [r̃] {q} can be performed in any order.
Another use of the column vector notation arises when taking derivatives of a function
with respect to a set of variables, or when taking the derivative of a scalar with respect
to a vector. Consider a vector {q} = [q1 q2 . . . qn ]T of dimension n, where the elements
q1 , q2 , . . . , qn are variables that are independent of each other, and a scalar S which is a
function of these variables, S = S (q1 , q2 , . . . , qn ). The derivative of S with respect to the
30 Applied Dynamics

vector {q} is defined as the n-dimensional row vector dS/d{q}, whose elements have the
form
 
∂S ∂S ∂S ∂S
= ... (2.20)
∂{q} ∂q1 ∂q2 ∂qn
The derivative of one column vector with respect to another can be obtained in a similar
fashion. Consider the column vector {v} of order m, where {v} = [v1 v2 . . . vm ]T , where
the elements of {v} are functions of q1 , q2 , . . . , qn . The derivative of {v} with respect to
{q} is a matrix of order m × n having the form
 ∂v1 ∂v1 ∂v1 
∂q1 ∂q2 . . . ∂q n
 ∂v2 ∂v2 ∂v2 
d{v}  ∂q ∂q2 . . . ∂qn 

=  1
(2.21)
d{q}  ...
 

∂vm ∂vm
∂q1 ∂q2 . . . ∂v
∂qn
m

For the special case when the scalar S is in quadratic form and expressed as S =
{q}T [D] {q}, where the elements of the matrix [D] are not functions of the variables
q1 , q2 , . . . , qn , the derivative of S with respect to {q} has the form
dS d{q}T [D] {q} T
= = {q}T [D] + {q}T [D] (2.22)
d{q} d{q}
and when the matrix [D] is symmetric we obtain
dS d{q}T [D] {q}
= = 2{q}T [D] (2.23)
d{q} d{q}

Example 2.2
Consider the scalar S = 3x2 + 4y 2 − 5xy and express it in terms of a symmetric matrix [D].
Writing the variables in vector form as {q} = [x y]T and taking the derivative of S with
respect to {q} gives
dS
{v}T = = [6x − 5y 8y − 5x] = 2{q}T [D] [a]
d{q}
from which it follows that
T
{v} = 2 [D] {q} = 2 [D] {q} [b]
and  
d{v} 6 −5
= 2 [D] = [c]
d{q} −5 8
Thus, the matrix [D] is  
3 −2.5
[D] = [d]
−2.5 4

2.5 Commonly Used Coordinate Systems

This section discusses four coordinate systems that are commonly used to describe motion.
One of these, rectilinear coordinates, involves unit vectors that are fixed in space, and the
other three are moving coordinate systems.
Kinematics Fundamentals 31

2.5.1 Rectilinear Coordinates

Z
P

$P
#
ZP
! Y
"
! XP
X

YP

FIGURE 2.4
Rectilinear coordinates.

The axes of a rectilinear coordinate system are fixed in direction. The unit vectors
along the coordinate axes are also fixed, and hence, their derivatives are zero. Consider
a coordinate system XY Z with unit vectors I, J, and K along the X, Y , and Z axes,
respectively. Another commonly used coordinate set is xyz with unit vectors i, j, and k.
Also, consider a reference point O and a point P , as shown in Figure 2.4. The position
vector rP , which describes the position of point P , has the form
rP = XP I + YP J + ZP K (2.24)
in which XP , YP , and ZP are the coordinates of point P , that is, their distance from the
origin O along the X, Y , and Z axes. To obtain the velocity of point P , denoted by vP ,
the above expression is differentiated with respect to time. Noting that the time derivative
of the unit vectors is zero,
drP
vP = = ẊP I + ẎP J + ŻP K + XP İ + YP J̇ + ZP K̇ = ẊP I + ẎP J + ŻP K (2.25)
dt
with the overdots denoting differentiation with respect to time. Similarly, the acceleration
of point P , which is denoted by aP , has the form
dvP
aP = = ẌP I + ŸP J + Z̈P K (2.26)
dt
The advantage of using a rectilinear coordinate system is its simplicity and the ease
with which components of the motion in other directions are viewed. The simplicity of
rectilinear coordinates, however, is also their disadvantage, as rectilinear coordinates provide
no information about the nature of the path that is followed.
Rectilinear coordinates are useful when components of the motion can be separated from
each other. A common application is projectile motion.

Example 2.3
A basketball player wants to shoot the basketball into the hoop. The player is at a distance
L from the basket and the basket is at a height h from the player’s chest, from where
32 Applied Dynamics

the player launches the ball. The player wants the ball to travel as a projectile and reach
a height of 5h/4 before it begins its descent towards the basket, as shown in Figure 2.5.
Calculate the initial velocity v0 and angle θ with which the player needs to launch the ball.

z A
B

!
h
vo "
h

x
L

FIGURE 2.5
Basketball on a trajectory.

Neglecting the aerodynamics, the accelerations of the projectile in the x and z directions
are
ax = 0 az = −g [a]
and the initial velocities in the x and z directions are
vx0 = v0 cos θ vz0 = v0 sin θ [b]
It follows that the expressions for velocity and acceleration can be written separately in
the x and z directions as
vx = v0 cos θ x = v0 cos θt
1 2
vz = v0 sin θ − gt z = v0 sin θt −
gt [c]
2
Point A is the point where the peak amplitude is reached. This happens at time tA . At
this point, the vertical velocity is zero, or vz = v0 sin θ − gtA = 0. Solving for the time tA
we obtain
v0 sin θ
tA = [d]
g
and the height reached at this point is
2
1 v02 sin2 θ

v0 sin θ 1 v0 sin θ 5
z (tA ) = v0 sin θ − g = = h [e]
g 2 g 2 g 4
Next, consider the time it takes for the ball to reach the basket, that is, point B. The
horizontal distance traversed, L, can be expressed as x = L = v0 cos θtB . Solving for the
time tB to reach B we obtain
L
tB = [f ]
v0 cos θ
Introduction of the above expression into the height at time t = tB gives the height at
time tB as  2
L 1 L
z (tB ) = h = v0 sin θ − g [g]
v0 cos θ 2 v0 cos θ
Kinematics Fundamentals 33

There are two unknowns, v0 and θ. The two equations that need to be solved are Equa-
tions [e] and [g]. The solution can be simplified by introducing the variables u = v0 cos θ
and w = v0 sin θ. Equation [e] can be rewritten as

1 w2 5 5
= h =⇒ w2 = gh [h]
2 g 4 2

Introducing this result into Equation [g] and rearranging leads to a quadratic equation in
terms of u r
5g 1 gL2
u2 − Lu + = 0 [i]
2h 2 h
which can be solved as
r r r ! r r r !
1 gL2 5 5 1 gL2 5 1
u = ± −2 = ± [j]
2 h 2 2 2 h 2 2

Note that there are two solutions. After calculating u and w, the next step is to solve
for the launch angle θ using the relation
w v0 sin θ
= = tan θ [k]
u v0 cos θ

from which the angle θ is obtained as θ = tan−1 (w/u). We can then introduce the value of
θ to any one of the expressions for u or w to find the launch speed.
This example can be used as a parametric study to determine the best options for
maximizing the possibilities of scoring a basket.

2.5.2 Normal-Tangential Coordinates

Normal-tangential coordinates take into consideration the properties of the path taken by
the moving body, which is extremely useful. On the other hand, normal-tangential coordi-
nates are not very useful in describing position.





 






 

FIGURE 2.6
Particle on a curved path.

Consider a particle moving along a curved path. The normal-tangential coordinate sys-
tem is a moving coordinate system attached to the particle. Two principal directions describe
the motion, normal and tangential. To obtain these directions, consider the position of the
particle after it has traveled distances s and s + ∆s along the path, as shown in Figure 2.6.
34 Applied Dynamics

The associated position vectors, measured from a fixed location, are denoted by r (s) and
r (s + ∆s), respectively. Define by ∆r the difference between r (s) and r (s + ∆s); thus,

∆r = r (s + ∆s) − r (s) (2.27)

As ∆s becomes small, ∆r and ∆s have the same length and become parallel to each
other. Further, ∆r becomes aligned with the tangent to the curve. The tangential direction
is defined as the direction tangent to the curve with the positive direction in the same
direction as the velocity. The unit vector in the tangential direction is defined as
∆r dr
et = lim = (2.28)
∆s→0 ∆s ds
The tangential direction is shown in Figure 2.7. The unit vector et changes direction as the
particle moves.

 

 










FIGURE 2.7
Normal and tangential directions.

The velocity is obtained by differentiating the displacement vector with respect to time.
Using the chain rule for differentiation gives
dr dr ds
v = = (2.29)
dt ds dt
Using the definition of et from Equation (2.28) and noting that the speed v is the rate
of change of the distance traveled along the path, v = ds/dt, the expression for velocity
becomes

v = vet (2.30)

The second principal direction is defined as normal to the curve and directed toward
the center of curvature of the path, and it is shown in Figure 2.7. This direction is defined
as the normal direction (n), and the associated unit vector is denoted by en . The center
of curvature associated with a certain point on a path lies along a line perpendicular to
the path. An infinitesimal arc of the curve in the vicinity of that point can be viewed as
Kinematics Fundamentals 35

a circular path, with the center of curvature at the center of the circle. The radius of the
circle is called the radius of curvature and is denoted by ρ. The two unit vectors introduced
above are orthogonal, that is, et · en = 0.
Differentiation of Equation (2.30) with respect to time gives the acceleration of the
particle a as

a = v̇ = v̇et + v ėt (2.31)

et(s)

et(s+ds) det en t

FIGURE 2.8
Infinitesimal change in tangential direction.

The derivative of et is needed to calculate the acceleration. To this end, consider Figure
2.7 and displace the particle by an infinitesimal distance ds along the path. The unit vectors
associated with the new location are et (s + ds) and en (s + ds). The center of curvature
remains the same as the particle is moved infinitesimally, so the arc length can be expressed
as ds = ρdφ in which dφ is the infinitesimal angle traversed as the particle moves by
a distance ds. Define the vector connecting et (s + ds) and et (s) by det , so that det =
et (s + ds) − et (s). From Figure 2.8, the angle between et (s + ds) and et (s) is small, so
that
ds
|det | ≈ sin dφ |en (s)| ≈ dφ = (2.32)
ρ
or
det 1
= (2.33)
ds ρ

The radius of curvature is a measure of how much a curve bends. For motion along a
straight line, the curve does not bend and the radius of curvature has the value of infinity.
For plane motion, using the coordinates x and y such that the curve is described by y = y(x),
the expression for the radius of curvature can be shown to be

1 d2 y/dx2
= h i3/2 (2.34)
ρ 2
1 + (dy/dx)

The absolute value sign in the above equation is necessary because the radius of curvature is
defined as a positive quantity. Considering the sign convention adopted above, the derivative
of the unit vector in the tangential direction becomes
det en
= (2.35)
ds ρ
36 Applied Dynamics

Using the chain rule, the time derivative of et becomes

det ds v
ėt = = en (2.36)
ds dt ρ

Introduction of this relationship to Equation (2.31) yields the acceleration as

v2
a = v̇ = v̇et + en (2.37)
ρ
The first term on the right in this equation is the component of the acceleration due
to a change in speed, referred to as tangential acceleration (at ). The second term is the
contribution due to a change in direction, referred to as the normal acceleration (an ). The
acceleration expression can be written as

a = at et + an en (2.38)

with at = v̇ and an = v 2 /ρ.

The normal and tangential directions define the instantaneous plane of motion, also
known as the osculating plane. The velocity and acceleration vectors lie on this plane. The
orientation of osculating plane changes direction (twists), as the particle moves.

Example 2.4—Road Curvature Design

Roads that change direction have to be designed with a curvature. The amount of curvature
depends on the maximum normal acceleration that a vehicle can have and not slide. When
designing a curving road, two important considerations are the amount of curvature and
the variation of the curvature as the vehicle enters and leaves a curve.1

b
an 0
an v 2/ b

b
b
2
an v / b
Center of
curvature
an 0

FIGURE 2.9
Connecting two roads by a quarter circle.

Suppose you are designing a connection between two roads. The roads are perpendicular,
as shown in Figure 2.9. One way is to fit a quarter circle (say, of radius b) to connect the
roads. The disadvantage of this design is that, even for a vehicle moving with constant
speed, the lateral acceleration will be zero before entering the curve and v 2 /b immediately
1 Banking of the curve is another important factor. We will address this issue in the next chapter, when

discussing kinetics.
Kinematics Fundamentals 37

after. This jump in acceleration may not be sustained by the friction between the tires
and the road surface. Further, it contributes to shock (or jerk, defined earlier as the time
derivative of acceleration) and causes discomfort.
A wiser curved road design is one where the radius of curvature changes gradually. Two
examples of such a curve are shown in Figure 2.10. The slow initial change in curvature
is usually compensated for by a higher curvature in the middle of the curve. The second
curve in Figure 2.10, where the vehicle first turns away from the curve (seems contrary to
intuition), is another approach to increase the smallest radius of curvature. This counter
steer action is also what a bicycle rider (or speed skater) does when taking a turn.

Countersteer
path
an = small

an = 0
an= large

an = small

an = 0

FIGURE 2.10
Improved road curvature design.

Example 2.5
The motion of a point is described in Cartesian coordinates as x (t) = 2t2 + 4t, y (t) =
0.1t3 +cos t, z (t) = 3t. Find the radius of curvature and normal and tangential accelerations
at t = 0.
To find the radius of curvature, we need to first calculate the normal and tangential
directions, as well as the speed and the acceleration components in terms of the normal and
tangential coordinates. The position, velocity, and acceleration vectors are

r (t) = 2t2 + 4t i + 0.1t3 + cos t j + 3t k

[a]

v (t) = dr (t) /dt = (4t + 4) i + 0.3t2 − sin t j + 3k

[b]
a (t) = dv (t) /dt = 4i + (0.6t − cos t) j [c]
The speed is
√
q
2 2
v = v·v = (4t + 4) + (0.3t2 − sin t) + 9 [d]
and the unit vector in the tangential direction can be written as et = v/v. At t = 0 the
velocity and speed are
p
v (0) = 4i + 3k v (0) = 42 + 32 = 5 [e]
38 Applied Dynamics

so that the unit vector in the tangential direction is

v 4 3
et = = i+ k [f ]
v 5 5
The value of the acceleration at t = 0 is

a (0) = 4i − j [g]

The tangential acceleration can be obtained from

at (0) = a (0) · et = (4i − j) · (4i + 3k) /5 = 3.2 [h]

and the normal acceleration becomes

an = an en = a − at et = (4i − j) − 3.2 (4i + 3k) /5 = 1.44i − j − 1.92k [i]

The magnitude of the normal acceleration is

√ p
an = an · an = 1.442 + 1 + 1.922 = 2.6 [j]

so the unit vector in the normal direction is

an 1.44i − j − 1.92k
en = = [k]
an 2.6
Taking the dot product with the unit vector in the tangential direction, we can confirm
that the two unit vectors are orthogonal to each other.
The radius of curvature can be calculated using

v2 52
ρ = = = 9.615 [l]
an 2.6

2.5.3 Cylindrical Coordinates

Cylindrical coordinates and their two-dimensional counterpart, polar coordinates, are pre-
ferred when motion is along a curved path, the distance of a point from an origin is of
interest, and one component of the motion can be separated from the other two. A common
use of polar coordinates is in orbital mechanics.
Consider a point P and an inertial coordinate system XY Z with point O acting as the
reference point from which P is observed, as shown in Figure 2.11a. The position of point
P can be described by first taking the projection of P onto the XY plane, denoted by
P 0 . The distance ZP from points P to P 0 is along the vertical direction, and it is one of
the parameters describing the motion. Next, draw a line from point O towards P 0 and call
the direction of this line the radial direction. The radial direction is also denoted as the r
direction and the associated unit vector along is denoted by er .
The distance R from points O to P 0 is the second parameter describing the motion. The
third parameter is the angle between the radial direction and X axis and is denoted by θ
and measured in radians. The unit vector along the radial direction, er , is

er = cos θI + sin θJ (2.39)

The transverse direction lies on the XY plane and it is perpendicular to the radial
direction. Its positive direction is along the direction of a positive rotation of θ. The unit
Kinematics Fundamentals 39

  


 


 


 



   
 





FIGURE 2.11
Cylindrical coordinates. a) Path of an object, b) the XY and rθ axes.

vector along the transverse direction, denoted by eθ , is along this direction and it obeys the
rule er × eθ = K. From Figure 2.11b, eθ is

eθ = cos θJ − sin θI (2.40)

The position of point P is expressed in cylindrical coordinates as

rP = XP I + YP J + ZP K = Rer + ZP K (2.41)

The unit vectors er and eθ change direction as point P moves. To obtain the velocity,
we need to differentiate the above equation

vP = ṙP = Ṙer + Rėr + ŻP K (2.42)

which requires the derivative of the unit vector in the radial direction. To calculate this
derivative, consider the projection of the motion onto the XY plane and that the particle
has moved to point Q, whose projection is Q0 . Consequently, the coordinate system has
moved by ∆θ, as shown in Figure 2.12.
The unit vectors of the new coordinate system are denoted by er (θ + ∆θ) and
eθ (θ + ∆θ) and related to er (θ) and eθ (θ) by

er (θ + ∆θ) = er (θ) cos ∆θ + eθ (θ) sin ∆θ

eθ (θ + ∆θ) = −er (θ) sin ∆θ + eθ (θ) cos ∆θ (2.43)

Using a small angle assumption of sin ∆θ ≈ ∆θ, cos ∆θ ≈ 1, and taking the limit as ∆θ
approaches zero, the derivatives of the unit vectors become

er (θ + ∆θ) − er (θ) der

lim = = eθ
∆θ→0 ∆θ dθ

eθ (θ + ∆θ) − eθ (θ) deθ

lim = = −er (2.44)
∆θ→0 ∆θ dθ
40 Applied Dynamics

 
 










FIGURE 2.12
Polar coordinate system moved by ∆θ.

The time derivatives of the unit vectors become

der dθ deθ dθ
ėr = = θ̇eθ ėθ = = −θ̇er (2.45)
dθ dt dθ dt
which, when substituted in the expression for velocity in Equation (2.42), results in

v = Ṙer + Rθ̇eθ + ŻP K (2.46)

The first term on the right side corresponds to a change in the radial distance and the
second term to a change in angle.
In a similar fashion we can find the expression for acceleration. Differentiation of Equa-
tion (2.46) yields

aP = v̇P = R̈er + Ṙėr + Ṙθ̇eθ + Rθ̈eθ + Rθ̇ėθ + Z̈P K (2.47)

Substituting in the values for the derivatives of the unit vectors and combining terms gives
   
aP = R̈ − Rθ̇2 er + Rθ̈ + 2Ṙθ̇ eθ + Z̈P K (2.48)

We can attribute a physical meaning to the acceleration terms. The first term, R̈, de-
scribes the rate of change of the component of the velocity in the radial direction. The
second term, Rθ̇2 , is the centripetal acceleration. This term is always in the negative radial
direction, as R is always positive. The term Rθ̈ describes the acceleration due to a change
in the angle θ. The next term, 2Ṙθ̇, is known as the Coriolis acceleration, named after the
French military engineer Gustave G. Coriolis (1792–1843). The Coriolis acceleration is due
to two sources. Both deal with a changing distance in a rotating system.
Cylindrical coordinates are suitable to use when one component of the motion, which is
selected as the Z (or z) direction, is separable from the other two.

Example 2.6
For the mechanism in Figure 2.13, the crank is at an angle γ = 30◦ and is rotating at the
rate of γ̇ = 0.2 rad/s, which is increasing by γ̈ = 0.1 rad/s2 . The crank causes the slotted
Kinematics Fundamentals 41





 

 
 


FIGURE 2.13
Crank and slotted link.

link to rotate. Using cylindrical coordinates, calculate ṙ and r̈ associated with point P on
the slotted link for the special case when b = a.
As in any kinematics problem, the analysis begins with examining the position, continues
on to velocity analysis, and then to accelerations. Because OP B is an isosceles triangle,
θ = γ/2, so that
γ̇ γ̈
θ̇ = θ̈ = [a]
2 2
The polar coordinates for the slotted link r, θ, the normal-tangential coordinates for the
crank t, n and the inertial coordinates X, Y are shown in Figure 2.14. The length r can
be shown to be
r = 2a cos θ [b]

 




 
 



FIGURE 2.14
Coordinate systems for crank and slotted link (for when a = b).

The velocity at the tip of the crank is

vP = vet = aγ̇et = 2aθ̇et [c]
where v = aγ̇. The velocity of point P in terms of polar coordinates is
vP = ṙer + rθ̇eθ [d]
Equating Equations [c] and [d] leads to the expression for ṙ. From Figure 2.14, the unit
vector in the tangential direction becomes
et = − sin θer + cos θeθ [e]
42 Applied Dynamics

so that
vP = 2aθ̇et = −2aθ̇ sin θer + 2aθ̇ cos θeθ [f ]
and considering Equation [c] gives

2aθ̇ cos θ γ̇
ṙ = −2aθ̇ sin θ θ̇ = = [g]
r 2
Note that ṙ can also be obtained by direct differentiation of Equation [b].
To find the second derivatives of r and of θ, we can either differentiate the above equation,
which yields
γ̈
r̈ = −2aθ̈ sin θ + 2aθ̇2 cos θ θ̈ = [h]
2
or obtain the acceleration terms by equating the normal-tangential and polar components
of the acceleration. In normal-tangential coordinates, the acceleration is

v2
aP = v̇et + en [i]
ρ

where v̇ = aθ̈ and ρ = a, so that

aP = aθ̈et + aθ̇2 en [j]

The acceleration components in polar coordinates are

   
aP = r̈ − rθ̇2 er + rθ̈ + 2ṙθ̇ eθ [k]

and the value for θ can be obtained by relating the components of the unit vectors in the
two coordinate systems.
It should be noted that when b 6= a, the solution becomes much more complicated from
an algebraic point of view, as r, θ, and γ are related by

r cos θ = a + b cos γ r sin θ = b sin γ [l]

This example shows that we can obtain solutions to kinematics problems either by
selecting coordinate appropriate systems or by finding algebraic relationships that describe
the geometry and differentiating these equations.

2.5.4 Spherical Coordinates

Spherical coordinates express position in terms of one displacement and two angular co-
ordinates. An important use of spherical coordinates is describing the position of a point
on Earth in terms of the point’s latitude and longitude. The configuration of spherical
coordinates is shown in Figure 2.15a.
There are several different conventions used to define the principal directions associated
with spherical coordinates. We use here the distance R of the point from a reference point
O, and two angles θ and φ, referred to as the azimuthal and zenith angles, respectively. The
parameter R here (total distance from reference point) is different from the R (distance
from reference point to projection onto the XY plane) used in cylindrical coordinates. The
azimuthal angle is the same as the polar angle in cylindrical coordinates.
The principal directions are referred to as the radial, azimuthal, and zenith. The radial
direction connects reference point O and point P , with the positive direction as outward.
Kinematics Fundamentals 43


 





   




  

FIGURE 2.15
a) Spherical coordinate system, b) side view.

The corresponding unit vector is denoted by eR , so that the position vector for P has the
form

rP = ReR (2.49)

To define the azimuthal and zenith directions, it is necessary to first select and orient an
inertial XY Z coordinate system. In Earth geometry, the equatorial plane is the XY plane
with the Z axis towards the north. Projection of point P onto the XY plane is denoted by
P 0 . Next, rotate the XY Z coordinates about the Z axis by the azimuthal angle θ to get an
xyz coordinate system, noting that the x axis goes through point P 0 . In Earth geometry,
the azimuth angle is the longitude. The zenith angle φ is defined as the angle that the Z
axis makes with the radial direction, as shown in Figure 2.15b. In Earth coordinates, the
zenith angle φ is known as the colatitude or 90◦ minus the latitude.
The unit vector in the radial direction can be expressed in terms of the xyz coordinates
as

eR = sin φi + cos φk = sin φi + cos φK (2.50)

Noting that the unit vector in the x direction is

i = cos θI + sin θJ (2.51)

the unit vector in the radial direction in terms of the XY Z coordinates becomes

eR = sin φ cos θI + sin φ sin θJ + cos φK (2.52)

As shown in Figure 2.16, the unit vector in the azimuthal direction is selected as similar
to its counterpart in cylindrical coordinates, the polar direction, so that

eθ = j = − sin θI + cos θJ (2.53)

We can show that eR · eθ = 0, so the two unit vectors are orthogonal. The unit vector
associated with the zenith angle satisfies the relationship

eφ = eθ × eR = cos φ cos θI + cos φ sin θJ − sin φK = cos φi − sin φk (2.54)

44 Applied Dynamics

 



 




FIGURE 2.16
Top view of spherical coordinate system.

The unit vectors eR , eθ and eφ form a mutually orthogonal set with eR and eφ lying on
the xz (or xZ) plane. We need to obtain the derivatives of the unit vectors associated with
the spherical coordinates in order to calculate velocities and accelerations. The procedure
is tedious and only the results are stated here:
ėR = sin φθ̇eθ + φ̇eφ ėθ = −θ̇ (sin φeR + cos φeφ ) ėφ = −φ̇eR + cos φθ̇eθ (2.55)

d
vP = ṙP = (ReR ) = ṘeR + Rθ̇ sin φeθ + Rφ̇eφ (2.56)
dt
   
aP = R̈ − Rφ̇2 − Rθ̇2 sin2 φ eR + Rθ̈ sin φ + 2Ṙθ̇ sin φ + 2Rφ̇θ̇ cos φ eθ

 
+ Rφ̈ + 2Ṙφ̇ − Rθ̇2 sin φ cos φ eφ (2.57)

Example 2.7
The airplane in Figure 2.17 is traveling in the Y Z plane and, at the instant shown, it
is executing a maneuver so that it is at the bottom of a vertical loop that has a radius of
curvature of 1500 m. The speed of the airplane is constant at 550 km/hr. A radar is tracking
the airplane. Using spherical coordinates and a radar at point O, find the values of R̈ and
φ̈.
Since the azimuthal angle θ is not of interest (but its derivative is), without loss of
generality it can be set equal to zero in Figure 2.17, so that the xyz and XY Z coordinate
systems coincide. Shifting the reference point to O, as shown in Figure 2.18, OP P 0 forms a
5-12-13 triangle, so
φ = tan−1 (12/5) = 67.38◦ . The unit vectors are
12 5 5 12
eR = I+ K eθ = J eφ = eθ × eR = I− K [a]
13 13 13 13
√
and the radial distance is R = 12002 + 5002 = 1300 m.
The next step is velocity analysis. The velocity of the airplane is v = 550J km/h, and
in m/s it is
km km 1000 m 1 hr
550 = 550 = 152.77 m/s [b]
hr hr 1 km 3600 s
Kinematics Fundamentals 45





 


 

 




 

FIGURE 2.17
Airplane tracked by radar.

eR
P

5 e
X
O 12 P'

FIGURE 2.18
Side view, with shifted axes.

The velocity is in the Y direction. The unit vectors in the radial and zenith directions
do not have components in the Y direction. Comparing with Equation (2.56) leads to the
conclusion that
12
v · eR = Ṙ = 0 v · eθ = Rθ̇ sin φ = 1300θ̇ = 152.77 m/s v · eφ = Rφ̇ = 0 [c]
13
Solving for θ̇ gives θ̇ = 152.77/1200 = 0.1273 rad/s.
The acceleration analysis is next. The aircraft is moving with constant speed on a curved
path. Considering normal-tangential coordinates, the only component of the acceleration is
in the normal direction (Z axis) and
v2 152.772 2
a = an = K = 15.56K m/s [d]
ρ 1500
The components of the acceleration in the radial, azimuthal, and zenith directions are
 
12 5 2
a · eR = 15.56K · I+ K = 5.985 m/s a · eθ = 0
13 13
 
5 12 2
a · eφ = 15.57K · I− K = −14.36 m/s [e]
13 13