Note:

vectors are written in boldface;

contrary to the common notation used in mathematics, a generic vector AB, with origin in A and endpoint
(tip) in B, will be denoted ( B – A ) instead of ( A – B ). In fact, if xA, yA, zA are the coordinates of the point
A and xB, yB, zB are the coordinates of the point B, the vector AB can be expressed as:
B B B

AB = ( B − A ) = i ( xB − xA ) + j ( yB − yA ) + k ( zB − zA )
1. KINEMATICS OF A MATERIAL POINT

1.1 Absolute reference systems

Let us consider an absolute reference system represented by a set of Cartesian axes, x-y (two-dimensional
system, 2D) or x-y-z (three-dimensional system, 3D). Let us denote by O the origin of the axes. Being the
reference system absolute the axes are fixed, that is the origin O does not move and the axes do not rotate.
For the sake of simplicity, first let us consider a two-dimensional case study. The positive versus of the axes
x and y is indicated by the versors (unit vectors) i and j. Since the axes of the reference system do not rotate
the versors i and j are constant. Therefore, their derivatives with respect to time are null:
di d j
=0 =0 (1.1.1)
dt dt

In the case of a point that moves in a 3D space the positive versus of the axes x-y-z is indicated by the
versors i , j and k.

1.1.1 Absolute position, velocity and acceleration of a material point

Let us consider a material point that moves in the x-y plane determined by an absolute reference system of
Cartesian axes. The loci curve of the time-varying position of the point, P(x,y), is the trajectory, s, which can
be a general curve or a straight line (Figure 1.1.1). In general, the trajectory can be expressed by means of a
function y = f (x), or by z = f (x, y), in the case of a point that moves in a 3D space.
Being the reference system fixed, the above mentioned trajectory s is called absolute trajectory of the
material point, expressed with respect to an absolute system of Cartesian coordinates.
For a given time t the point position is defined by the vector ( P − O ) :
=

( P −O) ( P − O ) eiϑ (P −O) ρ eiϑ (1.1.2)

where ( P − O ) , or ρ, is the magnitude of the vector ( P − O ) , while θ is the vector phase. Since the
position of P changes in the time, the two parameters ρ and ϑ are time-dependent functions. That is,
=

ρ ρ ( t ) and ϑ ϑ ( t ) .

Figure 1.1.1 Absolute trajectory, s, and position vector Figure 1.1.2 Absolute finite displacement of a particle
of a particle

The position vector ( P − O ) can be also expressed as:

(P −O) = i x + j y that is: ( P − O ) = i x(t ) + j y (t ) (1.1.3)

Since this position vector is defined with respect to fixed reference axes, x-y, the vector ( P − O ) represents
the absolute position of P.
If we consider two instants, t1 and t2, the point position changes from P1 to P2 and the finite displacement of P
is defined by the vector ( P2 − P1 ) (Figure 1.1.2). That is by the difference between the vectors ( P2 − O ) and
( P1 − O ) .

( P2 − P1 ) = ( P2 − O ) − ( P1 − O ) = i ( x2 − x1 ) + j ( y2 − y1 ) (1.1.4)

The instantaneous velocity v of the point can be determined by deriving the position vector ( P − O ) with
respect to time:
d (P −O) d ( x) d ( y)
v= =i +j = i x + j y (1.1.5)
dt dt dt

Since the vector ( P − O ) is the absolute position of the point, its derivative gives is the corresponding
absolute velocity. The two terms x and y , are the components of the velocity v along the directions x and y.
The magnitude of v is given by v = x 2 + y 2 while the phase α can be expresses as:

⎛ d y ⎞ ⎛dy dt ⎞ ⎛dy⎞
α = tan -1 ⎜ ⎟ = tan -1 ⎜ ⎟ = tan -1 ⎜ ⎟ (1.1.6)
⎝ d y ⎠ ⎝ dt dx⎠ ⎝dx⎠

Therefore, the phase α of the velocity vector v is the angle of the straight line that is tangent to the trajectory
in the considered point P. Then, the absolute velocity vector v is always tangent, in P, to the absolute
trajectory (Figure 1.1.4).

Figure 1.1.3 Absolute position of a particle Figure 1.1.4 Absolute velocity of a particle

The same vector v can be expressed also in this form:

v = v ei α = v e i α (1.1.7)

where the magnitude of vector v is v = v(t) and the phase α is a time-dependent function: α = α (t).
Let us denote t the unity vector that, at the time t, defines the direction and versus of the velocity vector v.
That is: t = 1 ei α = ei α = ei α ( t ) . Therefore, in general, the unity vector t depends on time, t: t = t ( t ) . If the
instantaneous value of the phase, α, of the unity vector changes, the corresponding angular speed, α or ω, is
given by: α = ω = d α d t . Then, the eq.(1.1.7) can be written as: v = v t .
The eq.(1.1.5) can be also rewritten as:

d ( P − O ) d ( ρ (t ) e i ϑ ( t ) )
v= = or v = ρ e i ϑ + ρ i ϑ e i ϑ (1.1.8)
dt dt

The imaginary index i is equivalent to the expression: i = e i ( ϑ + π 2)

.
Therefore, the eq.(1.1.8) can be rewritten as:
 v = ρ e i ϑ + ρ ϑ e i ( ϑ + π 2)
(1.1.9)

The two terms of this equation are vectors having amplitudes ρ and ρ ϑ , respectively, while the
corresponding phases are ϑ and ( ϑ + π 2 ) . That is, these two vectors are orthogonal and are the
components of the velocity vector v along the directions given by the angles ϑ and ( ϑ + π 2 ) .
That is, the direction of the absolute velocity vector v is t, as it is tangent to the absolute trajectory. However,
the same vector can be decomposed along two orthogonal vectors directed as the axes x-y, or directed as ϑ
and ( ϑ + π 2 ) .

Then, the absolute acceleration, a, of the material point can be obtained by evaluating the derivative of the
velocity vector v, with respect to time. That is:

d ( v ) d ( i x + j y ) d2 ( x) d2( y)
a= = =i + j = i 
x + j 
y (1.1.10)
dt dt d t2 d t2

The eq.(1.1.10) gives the projections of the acceleration a on the axes x, y.

Contrary to the velocity v, the acceleration vector a does not have a pre-established direction, with the
exception of some particular cases like that of a rectilinear trajectory.
Usually, it is worthwhile to project the vector a on the couple of orthogonal directions that are normal and
tangential to the absolute trajectory. These directions are identified by the versors n and t, respectively. The
components of the acceleration a projected on the normal and tangential directions can be obtained as
follows:
a = n an + t at (1.1.11)

Let us consider a short arc of the absolute trajectory evaluated about the current position of the material
point. In general this trajectory is curvilinear.
The terms an and at are positive or negative scalar quantities. Then, the eq.(1.1.10) can be rewritten as:

d (v) d (vt) d v d (t) d (v) d (t)

a= = = t+v = t+v (1.1.12)
dt dt dt dt dt dt

It is important to consider that t is a time-dependent unity vector: t = t ( t ) . Therefore, the derivative of the
unity vector t with respect to time must be evaluated.
For a given position ( P − O ) the centre of curvature, C, and the radius of curvature ρ * of the trajectory can
be evaluated (Figure 1.1.5). These two geometrical parameters can change in the time depending on the
shape of the trajectory. For a rectilinear trajectory the radius of curvature is infinite while a circular trajectory
has a constant radius ρ * .

Figure 1.1.5 Trajectory centre of curvature Figure 1.1.6 Changes in the tangent versor t
Now, let us consider the very small – e.g. infinitesimal – displacement ds of the point that occurs in an
infinitesimal time interval d t = ( t2 - t1 ) . Therefore, the point position changes from P1 to P2. During the time
interval d t the inclination angle of the radius of curvature ρ * is subjected to an infinitesimal change equal
to dϕ. The angular speed with which the segment (C − O ) rotates is given by:

dϕ
ω= (1.1.13)
dt

Owing to the infinitesimal length of the considered time interval also the length of the trajectory that the
point has travelled is so small that it can be approximated with a very short arc of a circumference having a
radius ρ * , equal to the radius of curvature of the trajectory. That is, during the infinitesimal time interval d t
the radius of curvature ρ * can be assumed to be constant. Figure 1.1.6 shows the two unity vectors t1 and t2,
that are tangent to the trajectory in P1 and P2. For the sake of clarity, the dimension of the infinitesimal angle
dϕ has been exaggerated in Figure 1.1.6.

Figure 1.1.7 Changes in the tangent versor t

If the origin of the unity vector t1 is shifted in P2 it is possible to point out the small vector dt that represents
the difference between the two unity vectors t1 and t2. That is: d t = ( t2 - t1 ) . The magnitude of the vector dt
is d t = 1 d ϕ . The vector dt represents the change of the unity vector t1 caused by the change of curvature
of the trajectory when the position of P changes from P1 to P2.
Being the interval d t infinitesimal, the direction of the vector dt coincides with that of the unity vector n, that
is normal to the trajectory in P, whereas the versus of dt is opposite to that of the unity vector n. Since the
infinitesimal rotation dϕ occurs in the infinitesimal interval d t, the derivative d ( t ) d t is given by:

d (t) d (ϕ ) d (t)
=− n that is: =−ω n (1.1.14)
dt dt dt

With regards to this, be aware not to confuse the versor t, which is a vector, with time t.
As said above, the radius of curvature ρ * is constant the time interval d t and the small displacements d s can
be expressed as:

d s = ρ*d ϕ (1.1.15)

Therefore, the magnitude of the velocity, v, can be written as:

dϕ
v = ρ* = ρ *ω (1.1.16)
dt

Then, the eq.(1.1.12) can be rewritten as:

 d (v) d (t) d (v) d (v) d (v) v2
a= t+v = t − vω n = t − ρ *ω 2 n = t− *n (1.1.17)
dt dt dt dt dt ρ

Therefore, by considering the eq.(1.1.17), the two components an and at of the absolute acceleration a,
directed in accordance with the unity vectors n and t are:

v2 d (v)
a n = − ρ *ω2 = − at = (1.1.18)
ρ *
dt

That is the acceleration vector a can be expressed by means of two components, at and an, which are oriented
along the directions tangential and normal, respectively, to the trajectory, evaluated for the current position
of the point. The two acceleration vectors at and an are called the tangential and normal components of the
acceleration a. That is, the two vectors at and an are tangential and normal to the absolute trajectory of the
material point.
In the case of a rectilinear trajectory, at least close to the current position P, the radius of curvature is infinite
and the normal component an of the acceleration nullifies.
Moreover, in accordance with the eq.(1.1.18), also in the case of a constant magnitude of the velocity v, the
acceleration a can be not null in all the cases for which the angular speed ω is not null and the trajectory is
not a straight line.
The normal component, an, of the acceleration a is also called centripetal. The versus of this acceleration
component is always towards the centre of curvature of the trajectory. For a given value of the velocity v, the
lower is the radius of curvature ρ the higher is the centripetal acceleration an.
In general, both components at and an are not null, though their instantaneous value can nullifies depending
on the shape of the trajectory and the characteristics of the point motion.
As said above, the velocity vector v is always tangent to the trajectory, while the acceleration vector a can be
always projected on the tangential and normal direction to the trajectory, at any time t.
Therefore, the acceleration a can be written as:

d (v) d (v) v2
a = i 
x + j 
y or: a= t − ρ *ω2 n = t− *n (1.1.19)
dt dt ρ

The acceleration a can also be obtained by deriving the eq.(1.1.7) with respect to time. That is:
d v iα d v iα
e + v α e (
i α +π 2)
a= e + i v α e i α = (1.1.20)
dt dt

with ω = α and v = ρ *ω . Therefore the eq.(1.1.20) can be rewritten as:

d v iα
e + ρ* ω ω e (
i α +π 2)
a= (1.1.21)
dt

The sign of the angular speed ω can be positive (counter-clockwise) or negative (clockwise). This determine
the sign of the sign of the last term of the eq.(1.1.21).

The same kinetic parameters v and a can also be determined evaluating the first order and second order
derivatives of the absolute position vector ( P − O ) . Be careful of the fact that in the following equations the
term ρ is the magnitude of the position vector ( P − O ) . That is:

v = ρ e i ϑ + ρ ϑ e i ( ϑ + π 2)
(1.1.22)

and

d 2 ( P − O ) d 2 ( ρ (t ) e i ϑ ( t ) ) d ( v ) d ( ρ e i ϑ + ρ i θ e i ϑ )
a= = = = (1.1.23)
d t2 d t2 dt dt
That is:
  e i ϑ + i ρϑ
a = ρ i ϑ + i ρϑ   e i ϑ + i ρϑe i ϑ + i ρϑ i ϑ e i ϑ =
(1.1.24)
  e i (ϑ + π
ρ i ϑ + 2 ρϑ 2)
+ ρϑe i (ϑ + π 2)
+ ρϑ 2 e i (ϑ +π )

The sum of the two velocity vectors on the right side of eq.(1.1.22) is always tangential to the trajectory of
the material point. The eq.(1.1.24) shows that the acceleration vector a can be decomposed into four vectors
that are orthogonal to each other. The phases of these vectors are: ϑ , ( ϑ + π 2) , ( ϑ + π ) . These
acceleration components are usually different from the above mentioned components at and an (see
eq.(1.1.19)). However, their sum gives the same acceleration vector expressed by the eq.(1.1.21).

The theoretical background above illustrated, mainly for a point that moves along a planar trajectory, can
easily be extended to a more general case in which the trajectory is defined in a 3D space.

1.2 Centripetal acceleration

It is important to emphasise that the normal component an of the absolute acceleration a is always
centripetal. That is its versus is always directed towards the centre of curvature of the trajectory. With regard
to this, it is important to take into account the sign of the angular speed ω with which the versor t, tangent in
P to the trajectory, rotates.
For instance, the material point shown in Figure 1.2.1 moves leftward on the trajectory s. The velocity of the
material point can be written as:

v = v ei α = v e i α (1.2.1)

while the acceleration can be expressed as:

d v iα d v iα d v iα
e + v α e (
2)
e + v α e (
i α +π i α +π 2)
a= e + i v α e i α = = (1.2.2)
dt dt dt

Owing to the trajectory curvature, the derivative of the angle α with respect to time, α , which coincides
with the angular speed ω, is positive, as its versus is counter-clockwise. Therefore, the angle α + π 2
indicates the versus of the centripetal component, an, of the acceleration a.
Conversely, the material point shown in Figure 1.2.2 moves rightward on the trajectory s. In this case the
derivative of the angle α with respect to time, α , which coincides with the angular speed ω, is negative, as
its versus is clockwise. That is:
d v iα d v iα d v iα
e + v α e (
2)
e + v α e (
i α +π i α +π 3 2)
a= e + i v α e i α = = (1.2.3)
dt dt dt

Therefore, the angle α + π 3 2 indicates the versus of the centripetal component, an, of the acceleration a.
Anyhow, also in this case the component an of the acceleration a is centripetal.

Figure 1.2.1 Particle moving leftward Figure 1.2.2 Particle moving rightward
1.3 Examples

Some very simple examples of the evaluation of the kinetic parameters of a material point are shown below.

At first a rectilinear trajectory has been considered. In this case the radius of curvature of the trajectory is ∞ .
Therefore, the component at of the absolute acceleration is null.

v = v ei α = v e i α = v t (1.3.1)

Then, a circular trajectory has been considered as second case study. In this case the radius of curvature of
the trajectory is constant. Therefore, the absolute velocity v and acceleration a of the material point can be
evaluated as:

v = ρϑ e i ( ϑ + π 2)

(1.3.2)
a = ρ ϑe i ( ϑ + π 2)
+ ρ ϑ 2 e i (ϑ +π )

Figure 1.3.1 Particle moving along a rectilinear trajectory Figure 1.3.2 Particle moving along a circular trajectory
1.4 Mobile reference systems

In the previous Section the basic kinetic parameters of a material point have been defined with respect to an
absolute Cartesian reference system, the axes of which do not rotate while the origin O is fixed. However,
the kinetic parameters of a material point can also be evaluated using a mobile reference system, that is a
Cartesian system of axes (x-y, x-y-z) whose origin can move in the 3D space while the single axes can rotate.
Let us consider a set of absolute Cartesian coordinates, x’-y’, whose origin, O’, is fixed while the axes does
not rotate. Then, let us consider a pair of Cartesian coordinates, x-y, whose origin is O. The point O can
move in the plane x’-y’, while the axes of this Cartesian reference system can rotate about the z axis
(Figure 1.4.1). In the end, let us denote by s the absolute trajectory of a material point, defined with respect
to the absolute reference system x’-y’. At time t, the direction of the x axis, evaluated with respect to the
horizontal axis x’, is given by the time-dependent angle α ( t ) (Figure 1.4.1).

Figure 1.4.1 Position vector with respect to a mobile Figure 1.4.2 Velocity vector with respect to a mobile
reference system x-y reference system x-y

The position of the origin O of the relative Cartesian system ( x-y ), defined with respect to the absolute
Cartesian system ( x’-y’ ), is given by the vector ( O − O' ) . The instantaneous velocity, vO, and acceleration,
aO, (Figure 1.4.2) of the origin O can be expressed as:

d ( O − O' ) d 2 (O − O' ) d ( v O )
vO = aO = = (1.4.1)
dt d t2 dt

The instantaneous values of the angular speed and angular acceleration with which the axes x-y rotate about
the z axis are given by the two vectors ω and ω , respectively. In accordance with common conventions, the
scalar value of the angular speed ω is positive if the rotation about the z axis is counter-clockwise. The same
standard is used to define the positive versus of the angular acceleration ω .
Then, the magnitude of the angular speed and angular acceleration of the mobile reference system, about the
z axis, are given by:

d [α (t )] d 2 [α (t )] d [ω (t )]
ω= ω = = (1.4.2)
dt d t2 dt

The general motion of the mobile reference system is roto-translational. However, pure translational
motions ( ω = ω = 0 ) or a pure rotation about the z axis ( v O = a O = 0 ) can occur.
If the x-y axes rotates, the unity vectors i and j are not constant terms as their orientation is time-dependent.
Therefore, when the position vector is differentiated with respect to time, the derivatives d i d t and
d j d t may give not null terms. In fact, the unity vectors i and j can be expressed as:

i = eiα j = e i (α + π 2)
(1.4.3)
By differentiating these expressions it is possible to obtain:

d i d (e i α )
= = i α e i α = ω e i ( α + π 2)
=ω j
dt dt
(1.4.5)
d j d (e i (α + π 2)
) d (i e i α )
= = = i i α e i α = − ω e i α = − ω i
dt dt dt

The position vector ( P − O' ) of the material point can be expressed as:

( P − O' ) = ( O − O' ) + ( P − O ) = ( O − O' ) + ( i x + j y ) (1.4.6)

where the vector ( O − O' ) is the position vector of the origin of the mobile reference system with respect to
the absolute reference system.

1.4.1 Instantaneous motion of the mobile reference system

The instantaneous motion (atto di moto) of the mobile reference system is generally a super-position of an
infinitesimal translation and an infinitesimal rotation (one of them may be null). If the origin O of the mobile
reference system is fixed, that is if the vector (O – O’) is constant, the instantaneous motion of the mobile
reference system is a pure rotation about a fixed axis passing through the point O. In this case the mobile
reference system is called rotating. Conversely, if the axes x-y do not rotate, while the vector (O – O’) is not
constant, the instantaneous motion of the mobile reference system is a pure translation. In this case the
mobile reference system is called translating. In the end, if the axes x-y rotate and vector (O – O’) is time-
variant, the mobile reference system is called roto-translating.
It is important to remark that the curve described by the origin O of a translating reference system can be
both a straight line and a curvilinear trajectory. The basic characteristic of such a mobile reference system is
only given by the fact that the axes x-y do not rotate at any time. Figure 1.4.3 shows some examples of
translating reference systems. Figure 1.4.4 shows an example of a rotating reference system: the origin O of
the axes x-y is fixed, in this case. In the end, Figure 1.4.5 shows an example of a roto-translating reference
system.
The instantaneous motion (atto di moto) of a rigid body, a mechanical system or a reference system, is also
defined as a vector field that associates the proper velocity vector with each element of the system.

Figure 1.4.3 Translational reference system: the axes do not rotate Figure 1.4.4 Rotational reference system:
the origin O does not move
 Figure 1.4.5 Roto-translational reference system: the origin O moves and the x-y axes rotate

For an observer who is moving together with the mobile reference system the material point describes a
relative trajectory that is different from the absolute one. Besides, if the material point P is virtually joined to
the mobile reference system and an infinitesimal motion – compatible with the actual instantaneous motion
of the reference system – is assigned to it, the material point describes a very short arc of trajectory –
infinitesimal – whose shape depends on the type of the instantaneous motion of the reference system, i.e.:
- an infinitesimal translation if the reference system is a translating coordinate system;
- an infinitesimal rotation if the reference system is a purely rotating coordinate system;
- an infinitesimal roto-translation if the reference system is a roto-translating coordinate system;
In the following this trajectory is called drag trajectory (traiettoria di trascinamento).
Therefore, for a material point, an absolute trajectory, a relative trajectory and a drag trajectory can be
defined whenever a mobile reference system is used for a kinematical analysis. Be aware of the fact that, in
general, the relative and drag trajectories must be determined at every time t.

1.4.2 Velocity

The absolute velocity v of the point P can be obtained by evaluating the derivative, with respect to time, of
the eq.(1.4.6). That is:

d ( P − O' ) d ( O − O' ) d ( P − O)
v= = + = v O + ( i x + j y ) + ( j ω x − i ω y ) (1.4.7)
dt dt dt

That is (by denoting with b ∧ c the cross-product between the vectors b and c):

v = vO + v r + ( ω ∧ ( P − O) ) = vO + v r + ( ω ∧ ( i x + j y ) ) (1.4.8)

This equation can be rewritten as:

v = v r + ⎡⎣ v O + ( ω ∧ ( P − O ) ) ⎤⎦ = v r + ⎡⎣ v O + ( ω ∧ ( i x + j y ) ) ⎤⎦ = v r + ⎡⎣ v t 1 + v t 2 ⎤⎦ = v r + v t (1.4.9)

The term vr is the relative velocity of P, that is the velocity evaluated with respect to the mobile reference
system x-y. This velocity vector is tangent to the relative trajectory of P, which is determined by the loci
described by the time-varying position P(x,y), defined with respect to the mobile reference system.

v r = ( i x + j y ) (1.4.10)

The second term, vt, of the eq.(1.4.9) is the drag velocity. The vector vt is the component of the absolute
velocity v that is caused by the instantaneous motion of the mobile reference system that has been
considered for the kinetic analysis. In general, the velocity vt is given by two terms that depend on the
translational motion and the rotation of the reference system. In fact:

v t = vO + ( ω ∧ ( i x + j y ) ) = v t 1 + v t 2 (1.4.11)
The drag velocity vt is tangent to the drag trajectory. Therefore, every velocity vector – absolute, relative,
drag – is always tangent to the corresponding trajectory.
In the case of a purely translating reference system the velocity vt is only given by the term vO of the origin
of the axes x-y. In fact, in this case the term ω is null. Conversely, in the case of a purely rotating reference
system the velocity vt is only given by the term:
v t =ω ∧ ( i x + j y ) (1.4.12)

where the term ω, which is a vector orthogonal to the plane x-y, is the instantaneous angular speed with
which the axes of the mobile reference system rotate. In the general case of a roto-translating reference
system both terms of eq.(1.4.11) can be not null. By combining the relative trajectory and the drag trajectory
one obtains the absolute trajectory of the point P.
The components of the velocity vector vt2, in the directions x-y-z, can be easily obtained evaluating the
determinant of the following 3 × 3 square matrix:

i j k
v t 2 = ω ∧ ( P − O ) = ωx ω y ω z = i ( ω y z − ω z y ) − j ( ω x z − ωz x ) + k ( ω x y − ω y x ) (1.4.13)
x y z

where ωx, ωy, and ωz are the angular velocity with which the mobile reference system rotates about the x, y
and z axes, respectively. In the example shown in Figures 1.4.1 and 1.4.2 the angular velocities ωx and ωy are
null as the motion of the material point P occurs in the x’-y’ plane.

1.4.3 Acceleration

The absolute acceleration a of the point can be obtained by evaluating the derivative, with respect to time, of
the eq.(1.4.9). That is:

d 2 ( P − O' ) d 2 ( O − O' ) d 2 ( P − O) d ( v O ) d ( i x + j y ) d ( j ω x − i ω y )
a= = + = + + (1.4.13)
d t2 d t2 d t2 dt dt dt

Substituting the expressions (1.4.5) into the eq.(1.4.13) we obtain:

a = a O + ( i  y ) + ( j ω x − i ω y ) + ( − i ω 2 x − j ω 2 y ) + ( j ω x − i ω y ) + ( j ω x − i ω y )
x + j 

a = aO + ( i  y ) + 2 ( j ω x − i ω y ) − ( i ω 2 x + j ω 2 y ) + ( j ω x − i ω y )
x + j  (1.4.14)

a = aO + ( i  y ) + 2 [ω ∧ ( i x + j y ) ] − ( i ω 2 x + j ω 2 y ) + [ω ∧ ( i x + j y ) ]
x + j 

Then, the eq.(1.4.14) can be rewritten as:

a = aO + ( i  y ) + 2 ( ω ∧ v r ) − ω 2 ( P − O ) + [ ω ∧ ( P − O ) ]
x + j 
(1.4.15)
a = aO + ( i  y ) + 2 ( ω ∧ v r ) + ⎡⎣ω ∧ (ω ∧ ( P − O ) )⎤⎦ + [ ω ∧ ( P − O ) ]
x + j 

a = a r + a t + aC (1.4.16)

where:
a r = ( i 
x + j 
y) (1.4.17)

{ } { }
a t = aO + ⎡⎣ω ∧ (ω ∧ ( P − O ) )⎤⎦ + [ ω ∧ ( P − O ) ] = a t 1 + a t 2n + a t 2 t = a t 1 + a t 2 (1.4.18)

a Cor = 2 ( ω ∧ v r ) (1.4.19)
The term aO represents the instantaneous acceleration of the origin, O, of the mobile reference system
evaluated with respect of the absolute reference system.
The term ar of the eq.(1.4.16) is the relative acceleration of the material point evaluated with respect to the
mobile reference system. This acceleration vector can be projected on a couple of orthogonal directions that
are normal and tangential to the relative trajectory sr of the point P. These two terms can be denoted by arn
and art, respectively.
The term at of the eq.(1.4.16) is the drag acceleration of the material point. In the case of a roto-traslating
reference system the two acceleration vectors a t 2n and a t 2 t are normal and tangent to the component of the
drag trajectory associated with the rotation only of the mobile coordinate system x-y-z. The acceleration
vector a t 2 t can be obtained evaluating the determinant of the following 3 × 3 square matrix:

i j k
at 2 t = ω ∧ ( P − O ) = ω x ω y ω z = i ( ω y z − ω z y ) − j ( ω x z − ω z x ) + k ( ω x y − ω y x ) (1.4.20)
x y z

where ω x , ω y , and ω z are the angular acceleration with which the mobile reference system rotates about
the x, y and z axes, respectively. In the example shown in Figures 1.4.1 and 1.4.2 the angular acceleration ω x
and ω y are null as the motion of the material point P occurs in the x’-y’ plane.
The acceleration vector a t 2 n can be also expressed as:

a t 2 n = ω ∧ ⎡⎣ω ∧ ( P − O )⎤⎦ = ω 2 ( P − O ) = ω 2 ( i x + j y ) (1.4.21)

This vector is the centripetal component of the acceleration at 2 . It is always directed from the point P to the
centre of curvature of the component of the drag trajectory associated with the rotation only of the mobile
reference system.
In the end, the term aCor of the eq.( 1.4.16) is the Coriolis’ acceleration vector. This term depends on the
instantaneous values of both the angular speed with which the axes of the mobile reference system rotate and
the relative velocity vr. The acceleration aCor can be obtained evaluating the determinant of the following
3 × 3 square matrix:

i j k
a Cor = 2 ( ω ∧ v r ) = 2 ωx ωy ωz (1.4.22)
vr x vr y vr z

Owing to the cross-product rules, this vector is orthogonal to the relative velocity vector vr.
As said above, each velocity vector is tangent to the respective trajectory (absolute, relative, drag).
Conversely, the direction of the acceleration vector is not necessarily tangent to the respective trajectory.
However, it can be always decomposed into two vectors, normal and tangent, respectively, to the
corresponding trajectory (absolute, relative, drag).

In the case of a purely translating reference system the drag acceleration at is only given by the term aO of
the origin of the axes x-y. Besides, also the Coriolis’ acceleration aCor is null. In fact, in this case, the terms
ω and ω are null. Therefore, in this case the expression of the absolute acceleration a is given by:
a = a r + aO (1.4.23)

Conversely, in the case of a purely rotating reference system, the acceleration aO is null while the drag
acceleration at is only given by the terms:

a = a r + ⎡⎣ω ∧ (ω ∧ ( P − O ) )⎤⎦ + [ ω ∧ ( P − O ) ] + a Cor (1.4.24)

The approach above described allows one to evaluate the absolute velocity and acceleration of a material
point P that moves along a 2D trajectory. Other methods, here not described for the sake of brevity, are
available in literature to perform this task as well as to determine the kinetic parameters of a material point
that moves in a 3D space.

1.4.4 Examples

…
…
2. KINEMATIC ANALYSIS OF RIGID BODIES

A rigid body moving in a 3D space has six degrees of freedom (d.o.f.s), that is the three components of the
translational displacements, x-y-z, evaluated along a Cartesian coordinate system and the three rotations,
θx-θy-θz, about the above mentioned axes x-y-z. Conversely, a rigid body moving in a plane (2D space) has
three degrees of freedom (d.o.f.s), that is the two components of the translational displacements, x-y,
evaluated along a 2D Cartesian coordinate system and the rotation θz about an axis orthogonal to the plane in
which the body moves.
Let us consider a rigid body (Figure 2.1) moving in a plane, for the sake of simplicity. The Cartesian x-y
reference system, having origin in O, is fixed. The straight line r is fixed to the rigid body. The position of
the rigid body can be completely determined by assigning the coordinates, x and y, of a generic point A of
the body and the angular position of the straight line r, with respect to a reference axis (e.g. horizontal).
In a general case study the rigid body can be subjected to a rotor-translating instantaneous motion, that, at
any time t, can be always decomposed into an infinitesimal (or finite) rigid translation and an infinitesimal
(or finite) pure rotation about an axis orthogonal to the plane x-y.
Let us consider two points, A and B of this rigid body (Figure 2.2). The distance BA = ( B − A ) is
constant. The absolute position vectors of the points A and B, defined with respect to a fixed Cartesian
reference system x-y are given by:

( A − O ) = ( A − O ) eiα ( B − O) = ( B − O) ei β (2.1)

The relative position of B, with respect to A, is given by the vector:

( B − A) = ( B − O ) − ( A − O ) = ( B − O ) e i β − ( B − O ) ei α ( B − A) = ( B − A) e i ϑ (2.2)

Therefore, absolute position of B, can be expressed as:

( B − O ) = ( B − A) + ( A − O ) = ( B − A) e i ϑ + ( A − O ) e i α (2.3)

That is:

( B − O ) = ( B − O ) e i β = ( B − A ) e i ϑ + ( i xA − j yA ) (2.4)

The coordinates xA ( t ) and yA ( t ) , along with the angle ϑ ( t ) , are the three degrees of freedom of the rigid
body (the motion of which can occur only in a 2D space). Let us denote vA and aA the vectors of the absolute
velocity and acceleration of A, evaluated at the time t. These two kinetic parameters are assumed to be
known, along with the distance B A and the angle ϑ .
The absolute velocity of B can be obtained by differentiating the eq.(2.4):

v B = ρ ϑ e i (ϑ + π 2)
+ v A = ρ ω e i (ϑ + π 2)
+ vA = vr + vt (2.5)

where ρ is the constant distance | BA | and ω is the instantaneous angular speed with which the rigid body
rotates, at the time t ( ω = ϑ ), see Figures 2.3 and 2.4.
This result can be also obtained considering a translational Cartesian reference system, the origin of which,
located in A, moves with velocity vA and acceleration aA at the time t.
With regard to eq.(2.5), the velocity vA coincides with the term vt while the term vr is the relative velocity of
B with respect to the mobile reference system. In fact, the rigid body is roto-translating, therefore, as the
distance | BA | is constant, the relative trajectory of B, with respect to an observer who is translating with the
mobile reference system with origin in A, is a short arc of circle whose centre is A and radius ρ. If the
angular speed ω is known, the eq.(2.5) allows one to evaluate the absolute velocity vector vB of any point of B

the rigid body.

The absolute acceleration of B can be obtained by differentiating the eq.(2.5):

aB = ρ ϑ e i (ϑ + π 2)
+ ρ ϑ 2 e i ( ϑ + π ) + aA = ρ ω e i (ϑ + π 2)
+ ρ ω 2 e i (ϑ + π ) + aA (2.6)
where ω is the instantaneous angular acceleration with which the rigid body rotates, at the time t ( ω = ϑ ).
By considering the translating coordinate system above mentioned, the eq.(2.6) can be rewritten as:
aB = ar + at
(2.7)
i ( ϑ + π 2) i (ϑ + π )
a r = a rt + a rn = ρ ω e +ρ ω e
2
at = aA

where ar is the relative acceleration of B, with respect to the mobile reference system and at is the drag
acceleration. If the angular acceleration, ω , of the rigid body is known, the eqs.(2.6-2.7) allow one to
evaluate the absolute acceleration vector aB of any point of the rigid body, see Figures 2.5 and 2.6.
B

The term ρ ω 2 e i ( ϑ + π ) is the centripetal component of the relative acceleration ar: this vector is normal to
the relative trajectory of B. Conversely, the term ρ ω e i ( ϑ + π 2 ) is the tangential component of the relative
acceleration ar: this vector is tangent to the relative trajectory of B.
If necessary, also the drag acceleration at can be decomposed into the two vectors atn and att , normal and
tangent, respectively, to the absolute trajectory of the point A.
Therefore, the absolute velocity and acceleration vectors vB and aB of a generic point B of a rigid body can
B B

be evaluated by assuming that the following parameters are known, at the time t:
- the absolute velocity vector vA of a point A of the rigid body;
- the absolute acceleration vector aA of a point A of the rigid body;
- the absolute angular speed ω of the rigid body;
- the absolute angular acceleration ω of the rigid body.

Figure 2.1 Kinematic analysis of a rigid body: basic Figure 2.2 Kinematic analysis of a rigid body:
parameters position vectors

Figure 2.3 Kinematic analysis of a rigid body: relative Figure 2.4 Kinematic analysis of a rigid body: absolute
velocity of the point B velocity of the point B
 Figure 2.5 Kinematic analysis of a rigid body: Figure 2.6 Kinematic analysis of a rigid body:
relative acceleration of the point B absolute acceleration of the point B

2.1 Instant Centre of Rotation of a rigid body

The instant centre of rotation, also called instantaneous centre and instant centre, is the point in a body
undergoing planar movement that has zero velocity at a particular instant of time. At this instant the velocity
vectors of the trajectories of other points in the body generate a circular field about this point which is
identical to what is generated by a pure rotation.
Planar movement of a body is often described using a plane figure moving in a two dimensional plane. The
instant centre is the point in the moving plane about which all other points are rotating at a specific instant of
time. The continuous movement of a plane has an instant centre for every value of the time parameter. This
generates a curve called the moving centrode. The points in the fixed plane corresponding to these instant
centres form the fixed centrode.

Figure 2.7 Instant centre of a moving plane Figure 2.8: Pole of a planar displacement

2.1.1 Pole of a planar displacement

The instant centre can be considered the limiting case of the pole of a planar displacement. The planar
displacement of a body from position 1 to position 2 is defined by the combination of a planar rotation and
planar translation. For any planar displacement there is a point in the moving body that is in the same place
before and after the displacement. This point is the pole of the planar displacement, and the displacement can
be viewed as a rotation about this pole.

2.1.2 Construction for the pole of a planar displacement

First, select two points A and B in the moving body and locate the corresponding points in the two positions;
see the illustration. Construct the perpendicular bisectors to the two segments A1A2 and B1B2. The
intersection P of these two bisectors is the pole of the planar displacement. Notice that A1 and A2 lie on a
circle about P. This is true for the corresponding positions of every point in the body.
If the two positions of a body are separated by an instant of time in a planar movement, then the pole of a
displacement becomes the instant centre. In this case, the segments constructed between the instantaneous
positions of the points A and B become the velocity vectors vA and vB. The lines perpendicular to these
B

velocity vectors intersect in the instant centre.

2.1.3 Pure translation

If the displacement between two positions is a pure translation, then the perpendicular bisectors of the
segments A1B1 and A2B2 form parallel lines. These lines are considered to intersect at a point on the line at
infinity, thus the pole of this planar displacement is said to "lie at infinity" in the direction of the
perpendicular bisectors. In the limit, pure translation becomes planar movement with point velocity vectors
that are parallel. In this case, the instant centre is said to lie at infinity in the direction perpendicular to the
velocity vectors.

2.1.4 Instant centre of a wheel rolling without slipping

Consider the planar movement of a circular wheel rolling without slipping on a linear road (Figure 2.9).
The wheel rotates about its axis M, which translates in a direction parallel to the road. The point of contact P
of the wheel with road does not slip, which means the point P has zero velocity with respect to point M.
Thus, at the instant the point P on the wheel comes in contact with the road it becomes an instant centre.
The set of points of the moving wheel that become instant centres is the circle itself, which defines the
moving centrode. The points in the fixed plane that correspond to these instant centres is the line of the road,
which defines the fixed centrode.
The velocity vector of a point A in the wheel is perpendicular to the segment AP and is proportional to the
length of this segment. In particular, the velocities of points in the wheel are determined by the angular
velocity of the wheel in rotation about P. The velocity vectors of a number of points are illustrated in
Figure 2.9.
The further a point in the wheel is from the instant centre P, the proportionally larger its speed. Therefore,
the point at the top of the wheel moves in the same direction as the centre M of the wheel, but twice as fast,
since it is twice the distance away from P. All points that are a distance equal to the radius of the wheel 'r'
from point P move at the same speed as the point M but in different directions. This is shown for a point on
the wheel that has the same speed as M but moves in the direction tangent to the circle about P.

Figure 2.9 Rolling wheel

2.1.5 Additional information

Here below, some additional information about the Instant Centre of Rotation is reported. Let us consider the
rigid body shown in Figure 2.10. The point A is moving along the absolute trajectory sA with the absolute
velocity vA, at time t. This vector is tangent to the trajectory sA (Figure 2.10). Then, let us consider a generic
point B of the rigid body that is moving along the absolute trajectory sB, at time t. The absolute velocity of B
B

is unknown but its direction coincides with the straight line tangent in B to sB (Figure 2.11).
B
 Figure 2.10 Motion of a rigid body in a 2-D space Figure 2.11 Motion of a rigid body in a 2-D space

Let us denote P0 the Instant Centre of Rotation (ICR) of the rigid body (in the absolute motion), at time t.
Being P0 a ICR, its absolute velocity is null: v P0 = 0 . By considering a translational coordinate system with
origin in P0 we have:

v A = v P0 + ω ∧ ( A − P0 ) and v B = v P0 + ω ∧ ( B − P0 ) (2.8)

That is:

v A = ω ∧ ( A − P0 ) and v B = ω ∧ ( B − P0 ) (2.9)

In accordance with the cross-product rules, the vector vA is orthogonal to the vector ( A − P0 ) , while the
vector vB is orthogonal to the vector ( B − P0 ) . Therefore, the ICR P0 is given by the intersection between
B

the straight lines passing through A and B, which are normal to the trajectories sA and sB (Figure 2.12).
B

Figure 2.12 Instant Centre of Rotation (ICR), P0 Figure 2.13 Absolute velocity of the point B

Therefore, when the absolute velocity vector vB of a generic point, B, of a rigid body must be evaluated, the
B

following two basic case studies may occur:

1) the known data of a kinematical problem are:
- the coordinates of the ICR position, P0;
- the instantaneous value of the angular speed , ω , of the body;
(in this case study the absolute trajectory of the point B, and then the direction of the velocity vB, are B

unknown);
2) the known data of a kinematical problem are:
- the absolute velocity vector vA of a point, A, of the body;
- the absolute trajectory of the point B, the absolute velocity of which must be evaluated.

In the case (1) the velocity vector vB is given by: v B = ω ∧ ( B − P0 ) .

B
In the case (2) the preliminary evaluation of the position of the ICR P0 must be carried out. This problem can
be solved by evaluating the coordinates of the intersection between the straight line, r1, that is orthogonal, in
A, to the velocity vector vA and the straight line, r2, passing from the point B, that is orthogonal to the
straight line tangent, in B, to the absolute trajectory of B (Figure 2.12). The intersection point between the
two straight lines r1 and r2 is the ICR P0. After having determined the position of P0 it is possible to evaluate
the angular speed ω of the rigid body from the following expression: v A = ω ∧ ( A − P0 ) .
Then, the absolute velocity vector vB can be obtained by: v B = ω ∧ ( B − P0 ) , (Figure 2.13).
B

As said above, the instant centre of rotation, P0, of a rigid body subjected to a planar motion is a point of the
plane that, for an infinitesimal time interval about the current time, can be considered rigidly connected to
the body itself. The point P0 can be internal or external to the rigid body. The instant centre of rotation can be
defined for the absolute motion of the body and for the relative motion as well. When the ICR is defined for
the absolute motion of body, its instantaneous absolute velocity, vP0, is null. Conversely, when the ICR is
defined for the relative motion of body, its instantaneous relative velocity, vr P0, is null.

2.1.5.1 Relative centre of rotation for two contacting planar bodies

If two planar rigid bodies are in contact, and each body has its own distinct centre of rotation, then the
relative centre of rotation between the bodies has to lie somewhere on the line connecting the two centres. As
a result since pure rolling can only exist when the centre of rotation is at the point of contact (as seen above
with the wheel on the road), it is only when the point of contact goes through the line connecting the two
rotation centres that pure rolling can be achieved. This is known in Involute gear design as the pitch point,
where there is no relative sliding between the gears. In fact, the gearing ratio between the two rotating parts
is found by the ratio of the two distances to the relative centre. In the example in Figure 2.14 the gearing
ratio is: γ = AD/BD.

Figure 2.14 Example of relative centre of rotation. Two bodies in contact at C, one rotating about A and the other
about B must have a relative centre of rotation somewhere along the line AB. Since the parts cannot interpenetrate the
relative rotation centre must also be along the normal direction to the contact and through C. The only possible
solution is if the relative centre is at D.

Consider the Instantaneous Centre of Rotation for the relative motion of the circular-shaped body hinged in
B, with respect to the beam hinged in A. The relative velocity of the contact point C, with respect to the
beam, is tangent to the beam in C. That is, it is parallel to the straight line passing through A and C. The
relative trajectory of point B, with respect to the beam, must be a circle of centre A and radius AB, as the
distance AB is constant. Therefore, the relative velocity of B, with respect to the beam, is orthogonal to the
straight line passing through AB. Thus, the relative ICR, D, of the circular-shaped body is the intersection of
the straight lines AB and CD, which are orthogonal to the above mentioned relative velocity vectors of
points B and C (with respect to the beam hinged in A).
The ICR, D, has a null relative velocity, in the relative motion of the circular-shaped body hinged in B, with
respect to the beam hinged in A. Then, if we consider a translational reference system with origin in D, we
can write:
vD = vr D + vt D (2.10)

That is, the absolute velocity of point D is the sum of the relative velocity vector and the drag velocity
vector. Being the relative velocity v r D null, we obtain: v D = v t D .
This means that the velocity of point D, assumed to be fixed to the beam coincides with the velocity of the
same point assumed to be fixed to the circular-shaped body.
Let us denote as ω 1 and ω 2 the angular velocity of the circular-shaped body and beam, respectively (see
Figure 2.14). Then, the absolute velocity of point D, assumed to be part of the circular-shaped body, can be
written as:

(
v D = ω1 ∧ D − B ) (2.11)

The absolute velocity of point D, assumed to be part of the beam, can be written as:

(
vD = ω 2 ∧ D − A ) (2.12)

Therefore we can write:

ω 2 = ω1
( D−B)
( D−A )
(2.13)

In the and, the relative velocity of point C, with respect to the beam, is given by:

(
vr C = ω r ∧ C − D ) (2.14)

where the relative angular velocity ω r can be obtained assuming to assign to the system a motion opposite to
that of the beam. This way, the beam remains fixed while the circular-shaped body describes a relative
motion, with respect to the beam, with the following relative angular velocity:
ω r = ω1 + ω 2 (2.15)

Figure 2.15 shows a further example of an instant centre of rotation in the relative motion. The disk is rolling
on the inclined plane of a moving wedge, at an angle α to the horizontal direction. The wedge is moving with
an absolute velocity v. The angular speed of the disk is ω. We assume that no slipping occur between the
disk and the supporting plane. The contact point H, between the disk and the plane, has a null relative
velocity.

Figure 2.15 Example of relative centre of rotation.

Consider a translational reference system connected to the wedge, that is a reference system that translate
with the wedge velocity v. The relative trajectory of point O, with respect to the mobile reference system, is
a straight line parallel to the inclined surface of the wedge. The relative velocity of point O is given by:

vrO = ω ∧ ( O − H ) (2.16)

Then, the absolute velocity is given by:

va O = v r O + v (2.17)
3. KINEMATICS OF A ROTATING DISK

Let us consider a purely rotating disk that is moving on a rectilinear guide. Let us denote by H the contact
point between disk and guide and by ω the disk angular velocity (Figure 3.1). Therefore, there is no slip
motion between the disk and guide. As a consequence of this, the relative velocity of the point H, with
respect to the guide, is null. The absolute velocity of the point H1, assumed belonging to the disk, is equal to
the absolute velocity of the point H2, assumed belonging to the guide. If the guide is fixed, the absolute
velocity of the point H is null. Therefore, the point H is the Instant Centre of Rotation (ICR) of the disk, at
time t. Then, the velocity of the disk centre, O, is given by:
vO = ω ∧ ( O − H ) (3.1)
The absolute trajectory of the point O is a straight line, parallel to the guide (Figure 3.1). The direction and
versus of the velocity vO is given by the versor t (Figure 3.2), the direction of which is constant.
Let us evaluate the velocity and acceleration vectors of a generic point A of the disc. By considering a
translating reference system, x-y, with origin in O we can write: v t = v O .
In fact, the drag velocity vt coincides with vO. Besides, the disk is a roto-traslating rigid body. Therefore, its
motion relative to an observer who is fixed with the translating coordinate system x-y is only a pure rotation
with angular speed ω. The trajectory of A, relative to the mobile reference system, is a short circular arc
whose centre and radius are O and (A – O), respectively. Therefore, the absolute velocity of A is given by:
v A = v t + v r = vO + ω ∧ ( A − O ) (3.2)
As H is the disc ICR, the velocity vector vA (Figure 3.4) is also given by:
vA = ω ∧ ( A − H ) (3.3)

Figure 3.1 Rotating disk Figure 3.2 Rotating disk: contact point

Figure 3.3 Rotating disk: relative velocity of point A Figure 3.4 Velocity of point A, according to Rivals’
with respect to the translating reference system x-y theorem and ICR method

The relative acceleration ar can be decomposed into two vectors atn and att that are normal and tangent,
respectively, to the relative trajectory of the point A, with respect to the mobile reference system. In
accordance with Rivals’ theorem, the absolute acceleration of A can be written as:
 aA = a t + ar = a t + ( arn + a rt ) = aO + ω ∧ ⎡⎣ ω ∧ ( A − O ) ⎤⎦ + ω
 ∧( A − O) (3.4)

The components of the acceleration vector of point A are shown in Figure 3.5. The absolute acceleration
vector of point A is shown in Figure 3.6.

Figure 3.5 Components of the acceleration

Figure 3.6 Absolute acceleration of point A
vector of point A

Beware of the fact that the ICR method cannot be used to evaluate acceleration vectors. Therefore, the
acceleration of A must me evaluated using a mobile reference system.