Some examples. Constraints and Lagrange Multipliers.

Physics 6010, Fall 2010

Some examples. Constraints and Lagrange Multipliers.
Relevant Sections in Text: 1.31.6
Example: Newtonian particle in different coordinate systems.
We have already noted that the Lagrangian
1 2
L = m~r V (~r, t),
2
will give the equations of motion corresponding to Newtons second law for a particle
moving in 3-d under the influence of a potential V . To see this use Cartesian coordinates:
1
L = m(x 2 + y 2 + z 2 ) V (x, y, z, t).
2
Using,
V
L
=
,
x
x
L
= mx,

x
the EL equation for the x coordinate is easily seen to be (exercise)

V
m
x = 0.
x

Of course, the y and z coordinates get a similar treatment. We then get (exercise)
m~r + V = 0.
Let us consider cylindrical coordinates,
x = cos ,

y = sin .

Can you write down the equations of motion following from F~ = m~a in cylindrical (, , z)
coordinates? It is completely straightforward using the EL equations a distinct practical
advantage of this formalism. We just need to express the kinetic energy (T ) and the
potential energy (V ) in terms of cylindrical coordinates and take the difference to make
the Lagrangian, from which the EL equations are easily computed.
To compute the kinetic energy we take the kinetic energy
1
T = m(x 2 + y 2 + z 2 )
2
1

Some examples. Constraints and Lagrange Multipliers.

and substitute
sin ,
x = cos

cos ,
y = sin +

to get
1
T = m( 2 + 2 2 + z 2 ).
2
To get the potential energy we just substitute into V (x, y, z, t) for x and y in terms of
and . The Lagrangian is of the form
1
L = m( 2 + 2 2 + z 2 ) V (, , z, t).
2
Caution:
The notation for the potential energy function is the usual one used by physicists,
but it can be misleading because, strictly speaking, it violates standard mathematical
notational rules. V (, , z, t) means the potential energy at the location defined by (, , z)
and at the time t. V (, , z, t) is not the function obtained by substituting x , y in
V (x, y, z, t), which a strict interpretation of the function notation would require. What we
are calling V (, , z, t) is in fact the function obtained from V (x, y, z, t) by the substitution
x cos , etc. For example, V (x, y, z, t) = x2 t corresponds to V (, , z, t) = 2 t cos2 .
The EL equations are (exercise)
m
2 +
m

V
= 0,

d 2
V
( ) +
= 0,
dt

V
m
z+
= 0.
z

Exercise: Does the 2 term in the radial equation represent an attractive or repulsive
effect in the radial motion?
Exercise: Repeat this computation for spherical polar coordinates.
Lagrangians for Systems
Often times it is useful to view a dynamical system as consisting of several subsystems.
For example, the solar system can be modeled as consisting of ten particles interacting gravitationally. In the absence of interaction, the Lagrangian L0 for the total (non-interacting)
system can be viewed as the sum of Lagrangians for each of its parts:
L0 = L1 + L2 + L3 + . . . .
2

Some examples. Constraints and Lagrange Multipliers.

Here L1 , L2 , etc. are the Lagrangians for the subsystems. For example, if we have a
system of (non-interacting) Newtonian subsystems each Lagrangian is of the form (for the
ith subsystem)
Li = Ti Vi .
Here Vi is the potential energy of the ith system due to external forces not due to intersystem interactions, which we are ignoring for a moment. It is easy to see that L0 correctly
describes the motion of the system of non-interacting systems through its EL equations.
This is because the EL equations for the k th system involve derivatives of L0 with respect
to coordinates and velocities of the k th system and this just picks out Lk from the sum
P
L0 = j Lj .
As a simple example, let us consider a system consisting of the planets, viewed as
non-interacting point particles. They all move in a central force field due to the sun which
is non-dynamical in this model. The Lagrangian for the k th planet, with position ~rk is of
the form
1
GM mk
2
Lk = mk ~r k +
,
2
rk
where M is the mass of the sun and G is Newtons constant.
Of course, non-interacting systems are an idealization and, ultimately, are of little
physical interest. Interactions are what makes the world what it is. One of the main
ways to mathematically represent interactions between the subsystems is to introduce a
potential energy function, V = V (q1 , q2 , . . . , t) which couples various degrees of freedom.
We then have the Lagrangian for the interacting system given by
L = L0 V.
The effect of this potential energy function in the Lagrangian is to couple the motion of
the subsystems. You can see this by noting the EL equations for a degree of freedom can
now, in general, depend upon the other degrees of freedom (exercise).
As an example, let us consider the Lagrangian for a pair of electons in a Helium atom.
We view the nucleus as fixed, with charge Q; it is part of the environment. The system
consists of the two electrons, each with mass m and charge q. The configuration space is
R3 R3 = R6 , and we label points in the configuration space with position vectors ~r1 and
~r2 . The Lagrangian is
1 2 qQ
q2
1 2 qQ
+ m~r 2

.
L = m~r 1
2
|~r1 | 2
|~r2 | |~r1 ~r2 |
The first two terms represent electron 1 moving in the Coulomb field of the nucleus.
Likewise for the next two terms regarding electron two. With these first 4 terms alone
the electrons orbit the nucleus independently and do not interact among themselves. The
3

Some examples. Constraints and Lagrange Multipliers.

last term represents the interaction between the electrons, which is Coulomb repulsion. It
is this term which couples the motion of the two electrons and makes the EL equations
somewhat complex, lacking an explicit solution.
The other principal way to mathematically represent interactions is via constraints. In
this scenario, the coupling between subsystems will typically occur via the kinetic energy
function. We will give a couple of examples in what follows.
Example: Double Pendulum
Consider a system consisting of two plane pendulums (pendula?) connected in series.
Dont even try to write down the equations of motion using Newtons second law! The
Lagrangian analysis is straightforward.
To begin with, we have two particles moving in a plane. We denote their x and y
positions via (x1 , y1 ) and (x2 , y2 ), where the origin of coordinates is placed at the fixed
point of the double pendulum. The masses are m1 and m2 . The motion of the particles
is constrained: the lengths are l1 and l2 ; pendulum 1 is attached to a fixed point in space
and pendulum 2 is attached to the end of pendulum 1. Mathematically we have
x21 + y12 = l12 ,

(x2 x1 )2 + (y2 y1 )2 = l22 .

These two constraints on the 4 cartesian coordinates leaves 2 degrees of freedom for this
system. As we have already mentioned, the configuration of the system is uniquely specified
once the angular displacement of each pendulum from (say) the vertical is specified. These
angles generalized coordinates for this system are denoted by 1 and 2 .
The kinetic energy for mass 1 is easily seen to be (exercise)
1
1
T1 = m1 (x 21 + x 22 ) = m1 l12 12 .
2
2
To find T2 we note that, defining l2 and 2 in the same way as for mass 1, we have
x2 = l1 sin 1 + l2 sin 2 ,
y2 = l1 cos 1 l2 cos 2 .
Along any curve we have (exercise)
x 2 = l1 cos 1 1 + l2 cos 2 2 ,
y 2 = l1 sin 1 1 + l2 sin 2 2 .
The kinetic energy of mass 2 is then (exercise)
1
T2 = m2 (x 22 + y 22 )
2
h
i
1
2
2
2
2

= m2 l1 1 + l2 2 + 2l1 l2 cos(1 2 )1 2 .
2
4

Some examples. Constraints and Lagrange Multipliers.

The potential energy for particle 1 is (exercise)

V1 = m1 gy1 = m1 gl1 cos 1
and for particle 2 we get (exercise)
V2 = m2 gy2 = m2 g (l1 cos 1 + l2 cos 2 ) .
The total potential energy is then
V = V1 + V2 .
All together, the Lagrangian for this system is (exercise)
1
1
L = (m1 +m2 )l1 12 + m2 l22 22 +m2 l1 l2 cos(1 2 )1 2 +(m1 +m2 )gl1 cos 1 +m2 gl2 cos 2 .
2
2
It is straightforward, albeit a little tedious, to compute the EL equations. Note that the
third term in the Lagrangian represents a coupling of the velocities of the two masses
through the kinetic energy. This is a consequence of interaction of the pendulums induced
by the constraints discussed above.

Example: Plane pendulum with moving support.

Consider a plane pendulum (mass m2 , length l) whose point of support is a mass
m1 which can slide horizontally. Let us compute the Lagrangian and EL equations. To
begin with, we have two particles moving in two dimensions with coordinates (x1 , y1 ) and
(x2 , y2 ). The motion is constrained; we have
y1 = 0,

(x2 x1 )2 + y22 = l2 .

Here we have set the x axis along the line upon which m1 moves. We then have (exercise)
(x1 , y1 ) = (x, 0)
and
(x2 , y2 ) = (x + l sin , l cos ).
Thus the generalized coordinates are the horizontal displacement x1 x for m1 and the
angle made with the vertical for the pendulum with mass m2 . The kinetic energy is
therefore (exercise)


1
1
2
2
2
2

T = m1 x + m2 x + 2lx cos + l .
2
2
5

Some examples. Constraints and Lagrange Multipliers.

The potential energy is (exercise)

V = m2 gl cos .
The Lagrangian is


1
1
L = (m1 + m2 )x 2 + m2 2lx cos + l2 2 + m2 gl cos .
2
2
Once again note how the constraints have coupled the motion via the kinetic energy.
The EL equations for x are (exercise)
(m1 + m2 )
x+

d
(m l cos ) = 0.
dt 2

The EL equations for are (exercise)

m2 l2 +

d
(m lx cos ) + m2 l(x g) sin = 0.
dt 2

Note that the x equation of motion implies

(m1 + m2 )x + (m2 l cos ) = constant.
We say that this quantity is a constant of motion or an integral of motion or that this
quantity is conserved by the motion.
Exercise: What is the physical meaning of this conserved quantity?
In both of our pendulum examples the Lagrangians have no explicit t dependence. An
example of a system with a t-dependent Lagrangian can be obtained from the previous
example as follows.
Suppose the point of support of the pendulum is forced to oscillate with amplitude A
and frequency :
x = A cos t.
Then the Lagrangian is (exercise)


1
1
L = (m1 + m2 )A2 2 sin2 t + m2 2lA sin t cos + l2 2 + m2 gl cos .
2
2
Note the explicit t dependence of the Lagrangian. There is now only a single degree of
freedom, . The first term in the Lagrangian is purely a function of time and will not
contribute to the EL equations (exercise); it can be dropped from L. The EL equations
for are now
m2 [l2

d
(Al sin t cos ) l(A sin t + g) sin ] = 0.
dt
6

Some examples. Constraints and Lagrange Multipliers.

As an exercise you can check that this equation correctly describes the reduction of the
original EL equations by the substitution x = A cos t.
Example: Charged particle in a Prescribed Electromagnetic Field
Two of the fundamental interactions allow themselves to be treated fruitfully using
classical mechanics: gravity and electromagnetism. Of course, we have in mind macroscopic systems here. (The other fundamental interactions strong and weak only operate microscopically and require a quantum treatment.) Here we will give a Lagrangian
formulation of the dynamics of a charged (test) particle in a given electromagnetic field.
While a fully relativistic treatment is certainly feasible, we will stick to a non-relativistic
(slow motion) treatment for simplicity.
We consider a particle with mass m and electric charge q moving in a given electro~ r, t), B(~
~ r, t). The equations of motion come from the Lorentz force law,*
magnetic field E(~
which asserts that the force F~ at time t on a charge q located at position ~r and moving
with velocity ~v is given by
~ r, t) + q ~v B(~
~ r, t).
F~ (~r, ~v , t) = q E(~
c
Here, of course,

~v = ~r = x i + y j + z k.
Thus the equations of motion for the curve ~r = ~r(t) are
~ r(t), t) + q ~r (t) B(~
~ r(t), t) = 0.
m~r(t) q E(~
c
It is not possible to find a Lagrangian whose EL equations correspond to the Lorentz
force law without introducing the electromagnetic potentials. Recall that 4 of the eight
Maxwell equations, the homogeneous equations,
~
~ + 1 B = 0,
E
c t
~ = 0.
B
~ so that
are equivalent to the existence of a scalar potential and a vector potential A
~
~ = 1 A ,
E
c t
and
~ = A.
~
B
* We use Gaussian units. The text uses mks units.
7

Some examples. Constraints and Lagrange Multipliers.

You can easily check that these relations lead to electromagnetic fields satisfying the homogeneous Maxwell equations (exercise). Conversely, given any electromagnetic field satisfying the homogeneous Maxwell equations, one can find a function (~r, t) and a vector
~ r, t) such that the above relations are satisfied.
field A(~
Remark: One issue that arises here is that of gauge transformations: For each configuration
of the electromagnetic field there are infinitely many potentials that can describe it. You
~ correspond to a given E
~ and B,
~ then so do
can easily check that if and A
0 =

1
,
c t

~0 = A
~ + ,
A
where (~r, t) is any function. This change in potentials is called a gauge transformation.
Because the potentials are not uniquely defined by the electromagnetic fields, and because
the effect of the electromagnetic field on matter is via the Lorentz force law involving
~ and B,
~ the potentials have no direct physical significance, e.g., one cannot measure
only E
by studying the motion of test particles.
In terms of the potentials, the equations of motion are (in an inertial reference frame)
!
~ 1
1

A
~ = 0.
~r A
m~r + q +
c t
c
We now show that the Lagrangian
1
q~
L(~r, ~r , t) = m(~r )2 q(~r, t) + A(~
r, t) ~r
2
c
yields these equations of motion as EL equations. To do this, we consider the EL equation
for x(t) and compare with the x component of the Lorentz force law. The y and z EL
equations are handled in an identical manner. We have (exercise)
~
L
q A
= q
+
~r ,
x
x c x
and

L
q
= mx + Ax ,
x
c
so that, on a curve ~r = ~r(t), (exercise)
d L
q
q Ax
= m
x + ~r Ax +
.
dt x
c
c t
Here Ax means to take the gradient of the function Ax . The EL equations are therefore
~
q
q Ax
q A
m
x + ~r Ax +
+q

~r = 0.
c
c t
x c x
8

Some examples. Constraints and Lagrange Multipliers.

Using (exercises)
()x =
~
A
t

!
=
x

,
x
Ax
,
t



~
~r A
~ = ~r Ax + A ~r ,
x
x
it is easy to see that the EL equation for x is the same as the x component of the Lorentz
force law (exercise).
There is a technical issue of interest here. The Lagrangian is built from potentials
~
(, A). As we have already pointed out, there are infinitely many potentials for any given
electromagnetic field. Thus there are infinitely many Lagrangians describing motion in
a single electromagnetic field. Each of these Lagrangians will yield the same Lorentz
~ B).
~
force law, which is built from (E,
Each of the Lagrangians can be related via a
gauge transformation of the potentials. Apparently, under a gauge transformation of the
potentials the Lagrangian changes in just the right way so that the EL equations do not
change. How does this happen? You will explore this in a homework problem.

Constraints
Often times we consider dynamical systems which are defined using some kind of restrictions on the motion. For example, the spherical pendulum can be defined as a particle
moving in 3-d such that its distance from a given point is fixed. Thus the true configuration space is defined by giving a simpler (usually bigger) configuration space along with
some constraints which restrict the motion to some subspace. We now give a systematic
treatment of this idea and show how to handle it using the Lagrangian formalism.
For simplicity we will only consider holonomic constraints, which are restrictions which
can be expressed in the form of the vanishing of some set of functions the constraints
on the configuration space and time:
C (q, t) = 0,

= 1, 2, . . . m.

We assume these functions are smooth and independent so that if there are n coordinates
q i , then at each time t the constraints restrict the motion to a nice n m dimensional
space. For example, the spherical pendulum has a single constraint on the three Cartesian
configruation variables (x, y, z):
C(x, y, z) = x2 + y 2 + z 2 l2 = 0.
9

Some examples. Constraints and Lagrange Multipliers.

This constraint restricts the configuration to a two dimensional sphere of radius l centered
at the origin. To see another example of such constraints, see our previous discussion of
the double pendulum and pendulum with moving point of support.
We note that the constraints will restrict the velocities:
d
C i C
C =
= 0.
q +
dt
t
q i
For example in the spherical pendulum we have
~r ~r = 0.
There are two ways to deal with such constraints. Firstly, one can simply solve the
constraints, i.e., find an independent set of generalized coordinates. We have been doing
this all along in our examples with constraints. For the spherical pendulum, we solve the
constraint by
x = l sin cos , y = l sin sin , z = l cos ,
and express everything in terms of and , in particular the Lagrangian and EL equations.
In principle this can always be done, but in practice this might be difficult. There is another
method in which one can find the equations of motion without having to explicitly solve
the constraints. This is known as the method of Lagrange multipliers. This method is
not just popular in mechanics, but also features in constrained optimization problems,
e.g., in economics. As we shall see, the Lagrange multiplier method is more than just an
alternative approach to constraints it provides additional physical information about the
forces which maintain the constraints.
Lagrange Multipliers
The method of Lagrange multipliers in the calculus of variations has an analog in
ordinary calculus. Suppose we are trying to find the critical points of a function f (x, y)
subject to a constraint C(x, y) = 0. That is to say, we want to find where on the curve
defined by the constraint the function has a maximum, minimum, saddle point. Again, we
could try to solve the constraint, getting a solution of the form y = g(x). Then we could
substitute this into the function f to get a (new) function h(x) = f (x, g(x)). Then we find
the critical points by solving h0 (x) = 0 for x = x0 whence the critical point is (x0 , g(x0 ))
This is analogous to our treatment of constraints in the variational calculus thus far (where
we solved the constraints via generalized coordinates before constructing the Lagrangian
and EL equations). There is another method, due to Lagrange, which does not require
explicit solution of the constraints and which gives useful physical information about the
constraints.
10

Some examples. Constraints and Lagrange Multipliers.

To begin with, when finding a critical point (x0 , y0 ) subject to the constraint C(x, y) =
0 we are looking for a point on the curve C(x, y) = 0 such that a displacement tangent
to the curve does not change the value of f to first order. Let the tangent vector to
C(x, y) = 0 at the point (x0 , y0 ) on the curve be denoted by ~t. We want
~t f (x0 , y0 ) = 0

where C(x0 , y0 ) = 0.

Evidently, at the critical point the gradient of f is orthogonal to the curve C(x, y) = 0.
Now, any vector orthogonal to the curve orthogonal to ~t at (x0 , y0 ) will be proportional
to the gradient of C at that point.* Thus the condition for a critical point (x0 , y0 ) of f
(where C(x0 , y0 ) = 0) is that the gradient of f and the gradient of C are proportional at
(x0 , y0 ). We write
f + C = 0,
for some . This requirement is meant to hold only on the curve C = 0, so without loss of
generality we can take as the critical point condition
(f + C) = 0,

C = 0.

This constitutes three conditions on 3 unknowns; the unknowns being (x, y) and . The
function is known as a Lagrange multiplier. In fact, if we artificially enlarge our x-y plane
to a 3-d space parametrized by (x, y, ) we can replace the above critical point condition
with
+ C) = 0,
(f
is the gradient in (x, y, ) space. You should prove this as an exercise.
where
To summarize: the critical points (x0 , y0 ) of a function f (x, y) constrained to a curve
C(x, y) = 0 can be obtained by finding unconstrained critical points (x0 , y0 , 0 ) of a
function
f(x, y, ) = f (x, y) + C(x, y).
We can do the same thing with our variational principle. Suppose we have an action
for n degrees of freedom q i , i = 1, 2, . . . , n:
Z t2
S[q] =

dt L(q(t), q(t),

t)
t1

where the configuation space is subject to m constraints

C (q, t) = 0,

= 1, 2, . . . , m.

* This follows from the basic calculus result that the gradient of a function is orthogonal to
the locus of points where the function takes a constant value.
11

Some examples. Constraints and Lagrange Multipliers.

Let the solutions to the constraints be given in terms of generalized coordinates sA , A =

1, 2, . . . , n m,
q i = F i (s, t),
i.e.,
C (F i (s, t), t) = 0.
The functions F i determine the graph of the solution set of C = 0 in the configuration
space. The correct equations of motion can be obtained by substituting the solutions
q i = F i (s, t) into the Lagrangian for q i , thus defining a Lagrangian for sA , and computing
the resulting EL equations for sA . Using the same technology you used in your homework
to study the effect of a point transformation on the EL equations (principally the chain
rule), it is not hard to see that the correct equations of motion are then


F i L
d L
= 0.

sA q i dt qi q=F (s,t)
i

F have the geometric meaning of (a basis of) (n m) tangent

We note that the functions s
A
vectors to the (n m)-dimensional surface C = 0. Thus the equations of motion are the
statement that the projections of the EL equations along the surface must vanish.

Now we introduce the Lagrange multiplier method. We consider a modified action,

Z t2
Z t2

S[q, ] =
dt L = S[q] +
dt (t)C (q(t), t),
t1

t1

in which we have added m new configuration variables , = 1, 2, . . . , m; these are the

Lagrange multipliers. The variation of the new action is


Z t2 
Z t2 
L
d
L
C

i
i

S =
dt

q .
q +
dt C +
q i dt qi
q i
t1
t1
are
The EL equations of motion coming from L
L
d L
C

+ i = 0,
i
i
dt q
q
q
which come from the variations in q i and also
C = 0,
which come from variations of . We have (n + m) equations for (n + m) unknowns. In
principle they can be solved to get the q i and the as functions of t.
What is the meaning of these equations? Well, the constraints are there, of course.
But what about the modified EL expressions? The EL equations you would have gotten
12

Some examples. Constraints and Lagrange Multipliers.

from L now have a force term, C

. The force term is geometrically orthogonal to
q i
the surface C = 0 in configuration space. This you can see from the identity (exercise)

0=

C F i

C
(F
(s),
t)
=
.

q i sA
sA

F represend n m vectors tangent to the surface defined by C = 0.) Thus

(Recall that s

A
is that the EL expressions coming from
the meaning of the EL equations coming form L
L no longer have to vanish, they simply have to be orthogonal to the constraint surface
since the equations of motion say that

L
d L
C

= i .
i
i
dt q
q
q
One physically interprets this force term as the force required to keep the motion on
this surface.
It is easy to verify that these modified equations, (n + m) in number, are equivalent
the correct (n m) equations obtained for sA earlier. Indeed, we have the m equations
of constraint. And, given this constraint, to say the EL expression coming from L is
orthogonal to C = 0 is the same as saying its projection tangent to the surface vanishes,
i.e.,




F i L
d L
F i
C

=
= 0,
q i
sA q i dt qi q=F (s,t) sA
which is precisely the content of the equations for the sA we obtained above.

Example: Plane pendulum revisited

Let us study the plane pendulum using Lagrange multipliers. We model the system as
moving in a plane with coordinates (x, y) subject the constraint
C = x2 + y 2 l2 = 0.
Without the constraint the Lagrangian would be simply
1
L = m(x 2 + y 2 ) mgy.
2
According to our general prescription for incorporating the constraint, we construct the
modified Lagrangian
= 1 m(x 2 + y 2 ) mgy + (x2 + y 2 l2 ).
L
2
13

Some examples. Constraints and Lagrange Multipliers.

with the configuration space parametrized

The critical points for the action built from L,
by (x, y, ), should give us the critical points along the surface C = 0. To find the critical
points we construct the EL equations as usual. We get
x2 + y 2 l2 = 0,
coming from the variation of , and
2x m
x = 0,

2y mg m
y = 0,

coming from the variations of x and y, respectively.

Here we can see more explicitly how the Lagrange multiplier defines a force term
beyond the gravitational force. This force of constraint represents the force of the rigid
pendulum arm upon the particle and is given by
F~constraint = 2x
x + (2y mg)
y.
The typical analysis of EL equations involving Lagrange multipliers can now be nicely
demonstrated. First, the three EL equations can be solved for (exercise)
=

m
(x
x + y y + gy) .
2l2

Next, differentiation of the constraint twice reveals:

C = 0

x
x + y y = (x 2 + y 2 ),

so that the multiplier can be solved for in terms of the original velocity phase space
variables:
m
= 2 (x 2 + y 2 gy).
2l
Substituting this result back into the EL equations for x and y we get the equations of
motion for x and y with the effect of the constraint physically, the tension in the rod
taken into account:
m
m
x = 2 (x 2 + y 2 gy)x,
l

m
m
y = 2 (x 2 + y 2 gy)y mg.
l

Note we never had to solve the constraint! Still, as a nice exercise you can check that,
after solving the constraint with x = l cos , y = l sin , these remaining 2 equations are
equivalent the familiar equation of motion for a plane pendulum, namely,
g
= sin ,
l
where is the angular displacement from equilibrium.
14

Some examples. Constraints and Lagrange Multipliers.

Using Lagrangian multipliers, the equations of motion for x and y tell us that the
pendulum moves according to a superposition of forces consisting of (i) gravity, (ii) the
force of constraint F~constraint needed to keep the mass moving in a circle of radius l.
This latter force is supplied by the Lagrange multiplier terms in the equation of motion.
Indeed, thanks to these Lagrange multiplier terms, the radial component of the net force
is (exercise)
~r ~
m
F = (x 2 + y 2 ),
l
l
which is the centripetal force, as it should be.
To summarize: Given a dynamical system with coordinates q i and Lagrangian L, we
can impose constraints C (q, t) = 0 by the following recipe.
(i) Add variables the Lagrange multipliers to the configuration space,
= L + C ,
(ii) Define a Lagrangian on the augmented velocity phase space L
for the q i and degrees of freedom.
(iii) Compute the usual EL equations from L
The resulting equations will include the constraints themselves as equations of motion
coming from variations of . The equations coming from the variations of the q i will have
extra terms involving the multipliers. For Newtonian systems these terms represent the
forces in the system which are necessary to enforce the constraints.
Thus the Lagrange multiplier method has distinct advantages over our previous approach in which we just solve the constraints at the beginning.. Namely, you do not have
to explcitly solve the constraints in order to compute the equations of motion, and the
equations of motion have additional physical information: the forces of constraint.

15