Multivariable and Vector Analysis: Wwlchen
Multivariable and Vector Analysis: Wwlchen
W W L CHEN
c
W W L Chen, 1997, 2008.
This chapter is available free to all individuals, on the understanding that it is not to be used for financial gain,
and may be downloaded and/or photocopied, with or without permission from the author.
However, this document may not be kept on any information storage and retrieval system without permission
from the author, unless such system is not accessible to any individuals other than its owners.
Chapter 7
PATHS
xxxxx
7.1. Introduction
In this chapter, we discuss paths in Rn ; in particular, we are interested in paths in R2 and R3 . Before
we give any formal definition, let us consider two examples.
t
(1, 0)
At time t = 0, a particle at the point (1, 0) C starts to move at constant speed along C in the
anticlockwise direction and returns for the first time to this initial position at time t = 2. It is easy to
see that at any time t [0, 2], the position of the particle may be given by (t) = (cos t, sin t). Here
we are interested in the function
Note that C = ([0, 2]) = {(t) : t [0, 2]} is the range of the function .
Example 7.1.2. Consider a particle moving away from the origin 0 = (0, 0, 0) at time t = 0 in the
direction of the unit vector u R3 with constant acceleration a, and hence speed ta at any given time
t 0. In this case, the distance of the particle from the origin at time t is given by 12 t2 a, and so its
position is given by (t) = 12 t2 au. Suppose that we trace the movement of this particle from t = 0 to
t = T . Then we are interested in the function
The range of this function is given by ([0, T ]) = { 21 t2 au : t [0, T ]}, and is a line segment joining the
origin 0 and the point 21 T 2 au.
Note that the functions (1) and (2) above do not just trace out curves. They also give the position of
the particles at any time within the time interval.
: [A, B] Rn ,
of the function is called a curve, with initial point (A) and terminal point (B). Suppose that for
every t [A, B], we have (t) = (1 (t), . . . , n (t)), where 1 (t), . . . , n (t) R. Then the functions
i : [A, B] R, where i = 1, . . . , n, are called the components of the path .
Remarks. (1) We usually write (t) = (x(t), y(t)) and (t) = (x(t), y(t), z(t)) in the cases n = 2 and
n = 3 respectively.
(2) Note the distinction between a path and a curve. Quite often, distinct paths may share the same
curve. For example, the three distinct paths
satisfy ([0, 2]) = ([0, 1]) = ([0, 1]) = C, the unit circle in R2 .
(3)xxxxx
Often, we refer to the path (t) without specifying the domain of definition of the function . This
is a convenient abuse of rigour.
Example 7.1.3. Consider a circular disc of radius r standing on a level surface. Let C denote the centre
of the disc, and let P denote a fixed point on the rim of the disc. Suppose that at time t = 0, the point
P touches the surface, and is therefore directly below the point C. For convenience, let us assume that
this point where the disc touches the surface at time t = 0 is the origin (0, 0).
P
v C
r
tv
The disc now starts rolling to the right at constant speed v. We now wish to describe the path taken
by the point P . Clearly the point C is at position (0, r) at time t = 0. Its position at time t is given by
(tv, r). Note next that the circumference of the disc is 2r, and so the disc will complete one revolution
at time t = 2r/v. It follows that the angular speed of the disc is v/r. Now let (t) denote the relative
position of P with respect to C. Clearly P rotates around C in a clockwise direction with angular speed
v/r, so it follows that
vt vt
(t) = r cos + , r sin + ,
r r
where R is a constant. Clearly (0) = (0, r), so that cos = 0 and sin = 1, whence = /2.
Hence
vt vt vt vt
(t) = r cos + , r sin + = r sin , r cos .
r 2 r 2 r r
vt vt
(t) = (tv, r) + (t) = tv r sin , r r cos .
r r
Clearly the point P touches the surface when t = 2k, where k is a non-negative integer. The image
curve of the path
is given below.
t 2 3 4
(t + h) (t)
lim
h0 h
exists for every t [A, B], with the obvious restriction to one-sided limits at the endpoints of the interval
[A, B]. In this case, the vector
d (t + h) (t)
0 (t) = (t) = lim
dt h0 h
is called the velocity vector of the path , and the quantity k0 (t)k is called the speed of the path .
Remarks. (1) Note that we have borrowed some terminology from physics. This is entirely natural, as
this area of mathematics is, to a large extent, motivated by the study of various problems in physics.
(2) Note that if the path is given by (t) = (1 (t), . . . , n (t)), then the velocity vector is given by
0 (t) = (01 (t), . . . , 0n (t)) and the speed is given by k(t)k = (|01 (t)|2 + . . . + |0n (t)|2 )1/2 .
(4) The velocity vector 0 (t) is a vector tangent to the path (t) at time t. If C is the curve of the
path (t) and 0 (t) 6= 0, then 0 (t) is a vector tangent to the curve C at the point (t) C.
Example 7.2.1. For the cycloid (t) = (t sin t, 1 cos t) described in Example 7.1.3, the velocity
vector is given by 0 (t) = (1 cos t, sin t). Note that 1 cos t = 0 implies that sin t = 0, so the velocity
is never vertical. The speed of the path is
This is minimum and zero when cos t = 1, when the point P touches the surface. The speed is maximum
when cos t = 1, when the point P is at the maximum height.
Example 7.2.2. To study the path (t) = (cos t, sin t, t) in R3 , we first of all consider the first two
xxxxx
components, and study the path (t) = (cos t, sin t) in R2 . This path describes a circle on the plane,
followed in the anticlockwise direction. The third component t describes an increase in height with time
if we think of the third component as the vertical component. It follows that if we consider the cylinder
x2 + y 2 = 1 in R3 , then the path (t) wraps round this cylinder in an anticlockwise direction with the
third component increasing if we look from above. The curve of the path (t) is called a helix.
y
1
x
1 1
1
The path has velocity vector 0 (t) = ( sin t, cos t, 1) and speed k0 (t)k = (sin2 t + cos2 t + 1)1/2 = 2,
so the path has constant speed.
Suppose that (t) is a differentiable path. We have already indicated that if 0 (t0 ) 6= 0, then it is a
vector tangent to the path at the point (t0 ). It follows immediately that
THEOREM 7A. Suppose that (t) is a differentiable path in Rn . Then the tangent line to the path at
the point (t0 ) is given by
Example 7.2.3. The equation of the tangent line to the helix (t) = (cos t, sin t, t) at the point (t0 ) is
given by
Suppose that t0 = 2. Then (2) = (1, 0, 2), and the tangent line becomes L() = (1, 0, 2)+(0, 1, 1).
Writing L() = (x, y, z), we have x = 1, y = and z = 2 + . It follows that the tangent line to the
helix at the point (1, 0, 2) is given by x = 1 and z = y + 2. Try to visualize this from the picture in
Example 7.2.2.
Example 7.2.4. The equation of the tangent line to the cycloid (t) = (t sin t, 1 cos t) at the point
(t0 ) is given by
Suppose that t0 = 2. Then (2) = (2, 0), and L() = (2, 0) + (0, 0) = (2, 0), clearly not the
equation of a line. Observe that since 0 (2) = (0, 0), Theorem 7A does not apply in this case. In fact,
the tangent line is vertical.
Example 7.2.5. Let us return to the helix discussed in Examples 7.2.2 and 7.2.3. Suppose that a
particle follows the helix from t = 0 to t = 2 and then flies off at constant velocity on a tangent at
t = 2. We wish to determine the position of the particle at t = 4. Note that the particle is at position
(2) = (1, 0, 2) when t = 2, with tangential velocity 0 (2) = (0, 1, 1). It follows that its position at
t = 4 must be given by
xxxxx
(2) + (4 2)0 (2) = (1, 0, 2) + 2(0, 1, 1) = (1, 2, 4).
t
1
and speed
Note that while the hypocycloid is a differentiable path, its curve has cusps. Note, however, that the
velocity and speed are zero at these cusps.
We state without proof the following two theorems. The proofs are not difficult, and follow by applying
the usual differentiation rules to the components.
THEOREM 7B. Suppose that (t) and (t) are differentiable paths in Rn . Suppose further that a(t)
and b(t) are differentiable real valued functions. Then
d
(a) ((t) + (t)) = 0 (t) + 0 (t);
dt
d
(b) (a(t)(t)) = a(t)0 (t) + a0 (t)(t);
dt
d
(c) ((t) (t)) = (t) 0 (t) + 0 (t) (t); and
dt
d
(d) ((a(t))) = a0 (t)0 (a(t)).
dt
The above represent the sum rule, scalar multiplication rule, dot product rule and chain rule respec-
tively. Note also the vector product rule below which is valid only in R3 .
THEOREM 7C. Suppose that (t) and (t) are differentiable paths in R3 . Then
d
((t) (t)) = (t) 0 (t) + 0 (t) (t).
dt
In this section, we are interested in calculating the length of the curve followed by a path. To motivate
this, note that the speed k0 (t)k of a path (t) is the rate of change of distance with respect to time.
Definition. Suppose that : [A, B] Rn is a differentiable path. The velocity differential is given by
Remarks. (1) The velocity differential describes an infinitesimal displacement of a particle following
the path . The arc length differential describes the magnitude of this infinitesimal displacement.
dx dy dx dy dz
ds = , dt and ds = , , dt
dt dt dt dt dt
2 2 !1/2 2 2 2 !1/2
dx
+ dy
dx
ds = dt and ds = + dy + dz dt
dt dt dt dt dt
respectively.
Definition. Suppose that : [A, B] Rn is a continuously differentiable path. Then the quantity
Z B
`= k0 (t)k dt
A
Example 7.3.1. The cycloid : [0, 2] R2 : t 7 (t sin t, 1 cos t) has arc length
Z 2 Z 2
`= k0 (t)k dt = (2 2 cos t)1/2 dt = 8.
0 0
Example 7.3.2. The helix : [0, 2] R3 : t 7 (cos t, sin t, t) has arc length
Z 2 Z 2
`= k0 (t)k dt = 2 dt = 2 2.
0 0
Example 7.3.3. The hypocycloid of four cusps : [0, 2] R2 : t 7 (cos3 t, sin3 t) has arc length
Z 2 Z 2 Z /2
`= k0 (t)k dt = 3| cos t sin t| dt = 12 sin t cos t dt = 6.
0 0 0
2. For each of the following paths, find the equation of the tangent line at the point (t0 ):
a) : [0, 2] R2 : t 7 (et , cos t) b) : [0, 1] R3 : t 7 (t3 , t2 , t)
3. A particle follows the path (t) = (sin et , t, 4 t3 ) in R3 from time t = 0 to time t = 1, and then
flies off at a tangent at constant velocity. Determine its position at time t = 3.