Chapter Ill: Moving reference frame kinematics

111.1 DEFINITIONS AND NOTATION

Consider the situation shown below where rigid body B is traveling through space,
undergoing both translation and rotation. A particle P is also moving through space where
P not necessarily attached to rigid body B.

body fvred observer

space fixed observer

p
I j
REFERENCE FRAMES
Space fvced reference frame : RA
k Body fixed reference frame: R8

COORDINATE SYSTEMS
Space fued coordinate system A : i j, k
Body fixed coordinate system B : i,J,k
;<T:>
w- L
X I

fa
·- REFERENCE FRAMES
As we describe the motion of this system, we will consider two frames of reference:
• Reference frame RA that is fixed in space. A space-fixed observer is attached to
this frame, as shown.
• Reference frame RB that is attached to rigid body B. A body-fixed observer is
attached to this frame, as shown. This reference frame has the same angular velocity
iiJ as the rigid body.

COORDINATE SYSTEMS
We will also consider two sets of coordinate axes:
• The space-fixed coordinate system A, having unit vectors of i, J and k.
Therefore, the unit vectors i , J and k have fixed orientations.
• The body-fixed coordinate system B, having unit vectors of i, J and k. Therefore,
the unit vectors l, J and k have an angular velocity of iiJ. As we have seen in the
last chapter:
 REFERENCE FRAME TIME DERIVATIVES
With multiple reference frames being used, we need to distinguish between a time

derivative with respect to the different reference frames. Let RA(!!.._) represent a time
dt ·

derivative with respect to the stationary reference frame RA, and let Rs (;) represent a

time derivative with respect to the moving reference frame Rs .

Consider, for example, the vector rPIQ in the above figure written in terms of its B-
components:
rPIQ =xi+ y} +zk
Taking a time derivative of this vector with respect to reference frame A gives:
RA[clrPIQ) dx~ dy dzk. di
- - =-1+-;+- +x-+y-+z-
dj dk
dt dt dt dt dt dt dt

dx • dy • dz •
=-i+-j+-k+x (i!Jxi') + y (i!Jxj') +z ( wxk' )
dt dt dt

•
dx • dy • dz • [ • • ']
=-i+-J+-k+wx xi+ yj+zk
dt dt dt

Rs cirPIQ ) - -
= [ -:it +wxrPIQ

where:
Rs [clrPIQ) dx dy
- - =-1+-1+-k dz,
dt dt dt dt

cir
= derivative of ___!!_SL with respect to reference frame B
dt
For a general vector p, we can write

- =Rs
RA(dp)
dt
(dp)
dt
--
- +wxp (1)

where iiJ is the angular velocity of reference frame B with respect to reference frame A .

Moving reference frame kinematics 1/l:2 ME 562 - cmk

•
 111.2 VELOCITY USING MOVING REFERENCE FRAMES
rt The position vector of point P shown earlier can be written as (see figure):
rP10 = rQ10 + rPIQ (2)
Taking a time derivative of this vector with respect to reference frame A gives:

where
·,(°';:al=·, [ °'~:
0
J+·, [°';:Q l (3)

RA (arP/0) = v-P = velocity

. of P as seen by the fixed observer (4)

•,[ °'~;o] = VQ = ,e/odty of Q as ,een by the f,xed ob,en,e, (5)

From equation (I), we know that:

RA(arPIQ]=RB (arPIQ]+wxr (6)

PIQ
where:

'• [ °';;Q ] = ve/odty of P as ,een by the body fixed ob,en,e, = (VP t, (7)

Substitution of equations (4)-(7) into equation (3) gives:

vP = vQ +(vPL, +wxrPIQ (8)

OBSERVATION
From above we saw that:
RB[ -
arPIQ dx dyldz •
- =-1+-1+-k
dt dt dt dt
From this, we see that the location of the origin Q on reference frame B does not affect

the value of R8 [ °';;Q l·Because ofthis, we have dmpped the subscript "Q" when

writffig '• [ °';;Q ] =(VP t 1

. Furthermore, this poffits out thatthe actual location of the

observer on the reference frame B does not influence the body-fixed observation of the
motion of P. Stated differently, all observers on reference frame B see the same velocity
of P. This is a subtle, but very important, observation.

Moving reference frame kinematics ll/:3 ME 562 · cmk

 III.3 ACCELERATION USING MOVING REFERENCE FRAMES
The fixed frame observation for the velocity of point P was shown to:
Vp = VQ +(vp t1 +w X rPIQ (8)
Taking a time derivative of this vector with respect to reference frame A gives:

·,[~; )='' [:;)+'A (d(~t, )+'' (d('"::P/Ql) (9)

where

RA ( ~ ; ) = ii P = acceleration af P as seen by the f,xed observer (10)

RA ( : ; ) = iiQ = accelera,;on of Q as seen by the fixed observer ( 11)

RA[d(vPt1)=R8 [d(vPL1)+wx(v ) (12)

dt dt P rel

RA[d(wxrP1Q)l
---'---....;...:.. = RA (dw)
- xr_ +cox
_ RA(arPiQ]
--
dt dt PIQ dt

=
RA (d-;ii-J xrPIQ+cox [Rs (ardt
CO - -
l l
PIQ - -
+coxrPIQ

= ii X rPIQ +OJ X (vp ),el +OJ X (w X rPIQ)

(13)
•
where

ii =RA ( :~) (14)

Recall that:

(vP)rel = dx l + dy) + dz k
dt dt dt
Therefore,

Rs(d(vPL1] =d2x d2y ~+ d2zk. = (-aP )

-1+-1 - (15)
dt dt2 dt2 dt2 rel

where ( ii P) rel = acceleration of P as seen by the body-fixed observer.

Substitution of equations (10)-(15) into equation (9) gives:

I
ii P= iiQ + (ii PLi+ ii x rPIQ + 2iv x (vPLi+ iv x (iv x rPIQ) (16)

Moving reference frame kinematics Il/:4 ME 562 - cmk

J
 OBSERVATION
From above we saw that:
2 2 2 2
(- ) = Rs[d rPtQ) d x~ d y~ d zkA
=--1+--;+--
rel dt 2 dt 2 dt 2 dt 2
From this, we see that the location of the origin Q on reference frame B does not affect
2
the value of Rs [ d rPI Q). Because of this, we have dropped the subscript "Q" when
dt 2
2
writing Rs d rPIQ ) = (Zip) . Furthermore, this points out that the actual location of the
( dt2 rel

observer on the reference frame B does not influence the body-fixed observation of the
motion of P. Stated differently, all observers on reference frame B see the same
acceleration of P. This is a subtle, but very important, observation.

Moving reference frame kinematics 1/1:5 ME 562 - cmk

 1/1.4 COORDINATE AXES AND REFERENCE FRAMES
In our derivations in both this chapter and Chapter II, we have used sets of "coordinate
axes" that are attached to either stationary objects (space-fixed) or moving objects (body-
fixed). In this chapter, we also introduced the concept of"reference frames" that are
attached to either stationary objects (space-fixed) or moving objects (body-fixed).
Superficially, it might appear that the concepts of coordinate axes and reference frames
are interchangeable. However, these two concepts are quite different. To see this,
consider the following simple example.

EXAMPLE
A disk rotates about a fixed shaft at O with a constant angular speed of .Q . A particle P
moves along a straight path with a constant speed of vp on a stationary surface. Three
observers are standing still on the disk at locations C, D and E. A set of coordinate axes
with unit vectors ( i ,J ,K) are attached to a fixed frame A, whereas a set of coordinate
axes with unit vectors (1 ,] J) are attached to moving reference frame B of the disk. At
the instant shown, the two sets of unit vectors are aligned with each other.
a) What are the fixed frame observations of the velocity and acceleration of P in
terms of the (i ,J,k) unit vectors? In terms of the (7,J ,.q unit vectors?

•
b) What are the moving frame observations of the velocity and acceleration of P in
terms of the (i ,J ,K) unit vectors? In terms of the (1,] J) unit vectors?

E moving frame B ]

p
D
I
'
''

'~ci
fixed frame A

,..:
, d

Moving reference frame kinematics lll:6 ME 562-cmk

•
 IV.5 OBSERVATIONS AND DISCUSSION
a) When using equations (8) and (16), one needs to first define the choice of
reference frame of the observer to be used. Based on this choice:
OJ and ii are the angular velocity and angular acceleration vectors,
respectively, for the observer.
( v p) rel and ( ii P t el are the velocity and acceleration vectors, respectively,
of point P as seen by the observer.
Note that ( vP) rel and ( ii P) rel are the same for ANY observer fixed to the
reference frame .
b) The vector v PIQ = vP - vQ is known as the "velocity of P relative to point Q" .

If a non-rotating observer moves with point Q, then vPIQ represents the

velocity of Pas seen by the observer at Q.
If the observer at Q is rotating, then the velocity of Pas seen by the
observer at Q is (as given by equation (8)):
(vPt1 =vPtQ-OJxrPtQ
where OJ is the angular velocity of the observer at Q.
Therefore, the term -OJ x rPIQ accounts for the difference between the
observed velocity of Pas seen by rotating and non-rotating observers at Q.
c) Similarly, the vector ii PIQ = ii P - iiQ is known as the "acceleration of P relative
to point Q".
If a non-rotating observer moves with point Q, then ii PIQ represents the
acceleration of Pas seen by the observer at Q.
If the observer at Q is rotating, then the velocity of Pas seen by the
observer at Q is (as given by equation (8)):
(iip t, = iiPIQ -ii x rPIQ - 20J x ( vp t, -OJ x (OJ x rPIQ)
where ii is the angular acceleration of the observer at Q.

Therefore, the terms -ii x rPIQ - 20J x ( iiPt, - OJ x (OJ x rPIQ) account
for the difference between the observed acceleration of P as seen by
rotating and non-rotating observers at Q.
d) As described in the last chapter, one must use care in determining the angular
acceleration vector:
- dOJ
a=-
dt
since some components of the reference frame angular velocity OJ are about
body-fixed axes, some are about space-fixed axes and some about intermediate
axes.

Moving reference frame kinematics 1/I:8 ME 562-cmk