Lagrangian mechanics: Introduction

In 1788, 100 years after Newton’s principia, an Italian mathematician

called Lagrange provided an alternative approach to solve problems in
mechanics, rather than Newtonian mechanics. This approach we now
call it Lagrangian Mechanics. This approach uses the energy approach
to solve problems in mechanics.
The Lagrange’s approach works just as easily for highly constrained
systems, non-inertial frames of reference and non-Cartesian
coordinates, and therefore makes light work of hard mechanics
problems. It delivers exactly one “F = ma” like second order equation of
motion for each generalized coordinate.
 Introduction
• The Lagrangian formulation of classical mechanics is based on energy
conservation. Its not a new theory but a new perspective.
• The essence of Lagrangian dynamics lies in its elegant and powerful
approach to describing the motion of particles and systems in
classical mechanics. Instead of working directly with forces and
accelerations, Lagrangian mechanics focuses on a quantity called the
Lagrangian, denoted by L, which is the difference between the kinetic
energy (T) and potential energy (V) of a system. The Lagrangian is
expressed in terms of generalized coordinates (𝑞𝑖 ), which uniquely
describe the configuration of the system.
 Degrees of freedom (DOF)
• This is the minimum number of independent coordinates required to
describe dynamics of a system [i.e., position of a single particle or a
system/collection of particles]
• It is represented by f.
• For a particle moving along a line (curved or straight) DOF is 1.
• For a particle moving along a surface (curved or plane) DOF is 2.
• For a particle moving in space DOF is 3.
• For N particles moving freely in d dimensional world f = Nd
where N = number of particles, dimension of particle
 Cont.
Example:

• Curved path

• Straight line
 Cont.
• Particle moving in space:

• Particle moving on a surface

 Cont.
• A particle moving along a latitude or longitude had DOF = 1

Generalisation
If there are N free particles moving in d dimensional world, then dof = Nd
• If there are K constraints or restrictions, then dof = Nd – K.
where k = number of constraints.
 Cont.
• Consider the following situation:
• d = 3, N = 1, k = 2 means it cant’ travel
• in y and z directions.
• So f = (1x3) – 2 = 3-2 =1
• f = 1: 2nd way of description
• y = 0, z = 0
 Cont.
• Another way of description
 Worked Example 1
• A block slides on a wedge which is moving on a horizontal surface
[motion is confined in vertical plane]. What is dof of the system?
 Cont.
Solution
• f = Nd – k : N = 2, d = 2, k = 2
• f = (2x2) – 2: so f = 2
 Worked Example 2
Two particles are moving on surface of a sphere. Find degree of
freedom.
• Constraints: particle moving on
surface anywhere but distance
from the center is always r.
f = Nd – k: N =2, d = 3, K = 2
f = (2x3) – 2
f=6–2=4
 Worked Example 3
Two particles are connected by a spring:
• Find the dof.
Solution
• The spring does not put any condition
(i.e. constraints)
So f = 4
 Worked Example 4
Suppose a free particle is moving on a surface plane: Find dof.
Solution
• For surface: d = 3, constraints k = 1
Therefore by f = Nd – k ⇒ f = 1x3 – 1 = 2
If d = 2, k = 0 , since f = Nd – k then f = 1x2 – 0 = 2
 Worked Example 5
For particle moving along a line
• A line is formed by intersection of two planes. So if a particle moves
along a line, then it can be assumed to be in contact with two planes.
• Therefore k = 2, d = 3, N = 1
• f = Nd – k ⇒ f = 1x3 – 2 = 1.
• Formula f = Nd – k does not work if N > 4 and there is constraint
among all particles.
 Worked Example 6
1. Five free particles on a sphere:
N = 5, d = 3, k = 5, so f = Nd – k ⇒ f = 5 x 3 – 5 = 10.
2. Five particles are rigidly connected to each other
and the system moves in space:
• f = Nd – k, f = 5x3 – 10, f = 5
5! 5𝑥4
where: k = 5𝐶2 = = = 10
3!2! 2
but this is not a correct answer.
 DOF of a rigid body
• The DOF of a rigid body in d dimension world is given by:
𝑑(𝑑+1)
• f=
2
𝑑(𝑑+1)
• If it has k external constraint then f = -k
2
𝑑(𝑑+1)
• DOF of a free rigid body in 3D world: f = for d dimension
2
• For d = 3 , f = 6 only for 3 dimension (3 translational, 3 rotational).
• DOF of a rigid body with one fixed point in 3D world, then translation
will not be possible, and DOF = 3 (rotational)
 Cont.
• DOF of freedom of rigid body rotating about a fixed axis = 1 (only one
angle is needed)
Question
• A sphere is moving on a flat surface. What is its degree of freedom?
Response:
𝑑(𝑑+1) 3𝑥4
• f= -k ⇒ f= - 1 = 5
2 2
 Worked example 7
• A block which is sliding on a wedge which is free to move on a
horizontal surface [motion is confined in vertical plane]. What is dof
of the system?
Response
• Dof = 1 + 1 = 2
• A rigid body can be considered
as a particle if there is no
rotation
 Worked Example 8
• If a sphere is moving on a plane and an insect is crawling on it. What
is the dof of the insect?
• f = Dof of insect on stationary sphere + DOF of plane
=2+5=7
In summary
𝑑(𝑑+1)
For f = then 3-D : f = 6, 3 [rotation, 3 translation]
2
for 2-D then f = 3, [1 rotation, translation]
for 1-D then f = 1 [translation]
 Constraints (Conditions)
Geometrical condition on coordinates
• This means there is a relation between coordinates.
Types of constraints
• (i) Holonomic constraints (complete) – relation between coordinates
is in form of algebraic equation e.g., 𝑥 2 + 𝑦 2 = 𝑙 2
• (ii) Non-holonomic constraints
• If relation between coordinates is differential equation then it must
be reducible to algebraic equation
• e.g., x dy + y dx = 0 : then ‫ = 𝑦𝑥 𝑑 ׬‬0 ⇒ xy = constant (algebraic
equation)
 Holonomic Constraints
Example: simple pendulum
• The variable coordinates are 2 which are
x and y.
They are related by the equation 𝑥 2 + 𝑦 2 = 𝑙2 .
This equation forms a constraint. It governs the
the motion of the pendulum.
The equation is called the constraint equation
• 𝑦 2 = 𝑙2 - 𝑥 2 , ⇒ meaning if you know x we can find y. So y is dependent on x, or
vice versa. On this pendulum,
DOF = [number of related variable coordinates] - [no. of constraint equations]
(for holonomic)
 Cont.
Spherical pendulum
• The number of related
coordinate variables
is 3, which are (x, y, z)
• Constraints equation
is: 𝑥 2 + 𝑦 2 + 𝑧 2 = 𝑙 2
(holonomic) and is 1
• Degrees of freedom becomes:
DOF = 3 - 1 = 2.
 Cont.
Block and Wedge: DOF = 3 – 1 = 2
• Here h and θ are fixed.
ℎ−𝑦1
Tan θ = :
𝑥1 −𝑥2

h - 𝑦1 = (𝑥1 - 𝑥2 ) tan θ

𝑦1 = h – (𝑥1 − 𝑥2 ) tan θ

This is the constraint equation

 Non Holonomic constraints
• Constraints equations are either inequality or non integrable
differential equations.
• It does not reduce degree of freedom.
• Example: a particle (e.g., a fly) moving inside a sphere.
• The non holonomic constraint equation
becomes: √(𝑥 2 + 𝑦 2 + 𝑧 2 ) ≤ R
• Thus, d = 3, N = 1, k = 0
• DOF = Nd – k ⇒ (1 x 3) - 0
• DOF = 3
 Cont
• Here there is one constraint ⇒ the fly cannot go out of the sphere,
but it is non holonomic.
• Example: Disc rolling on a
plane (xy-plane)
• The condition for rolling:
𝑑∅
• v=r𝜔⇒v=
𝑑𝑡
𝑑𝑦 𝑑∅
• 𝑣𝑦 = v cos θ : = R cos θ
𝑑𝑡 𝑑𝑡
• dy = R d∅ cos θ……………………1
 Cont.
• 𝑣𝑥 = - v sin θ, thus:
dx = - R d∅ sin θ …………………………….2
we divide 1 by 2
𝑑𝑦
= - cot θ
𝑑𝑥
So ‫ = 𝑦𝑑 ׬‬- ‫ ׬‬cot 𝜃 dx
This is the differential of x and y, with θ as a variable. So it cannot be
integrated. Therefore rolling is a non holonomic constraint when
motion is along a curved line. It is holonomic when motion is along a
straight line.
 Types of holonomic and non holonomic
constraints
• There are two types of these constraints:
• (i) Holonomic and Scleronomic (rigid)
• Time independent – the constraint
equation does not contain time
explicitly.
• Example: bead sliding on a fixed rod.
𝑦
• Constraint equation is: tan θ = , so y = x tan θ. This equation does
𝑥
not contain time.
 Cont.
• DOF = No. of variables – No. of constraint equations: DOF = 2 – 1
DOF = 1.
(ii) Rheonomic constraint- (Non rigid) – it is time dependent
Constraint equation completely contains time. An example is a bead
sliding on a rotating rod:
𝑦
Equation of constraint will be: tan θ =
𝑥
So, y = x tan (𝝎t) [where θ = 𝜔t)
⇒ it contains time
The equation is Holonomic and Rheonomic
 Cont.
• Example: particle moving inside expanding sphere. The equation of
constraint is:
• 𝑥 2 + 𝑦 2 + 𝑧 2 ≤ 𝑅𝑜 + vt

• This equation is Non-Holonomic and

Rheonomic
 Cont.
• Convenient coordinates used simplify problems are called Generalised
Coordinates.
• In most cases number of generalised coordinates is equal to degrees
of freedom.
Lets consider a particle moving
along a curved line (fixed).
What is the dof of the particle?
Solution: N = 1, k = 1, d = 3
f = Nd – k , f = (1x3) – 1 = 2
 Cont.
• However, this answer is not true here because a line is formed by
intersection of two planes.
• So dof of one plane is 1, and for another plane is 1. So total
constraints is 2.
• So DOF = f = Nd-k = (1 x 3) – 2 = 1
• If a particle moves in a constrained circle
• (𝑥 − 𝑎)2 + (𝑦 − 𝑎)2 = 𝑅2
• D = 3, k = 2, N = 1, DOF = Nd – k
• f = (1 x 3) – 2 = 1
 Cont.
• Note: If a particle moves along a fixed line, (it may be a curved line) or
a straight line, DOF = 1, because a line is formed by intersection of
two planes.
 Terminology in Lagrangian mechanics

Generalised coordinates
• It is convenient coordinates used to simplify a problem
• It may be cartesian, plane polar, spherical polar, or some other
variables
• The symbol q is used to represent generalised coordinates
Assumptions
• All generalised coordinates are independent of each other.
• i.e.,
 Cont.
Generalised velocity
• It is the total time derivative of generalised coordinates.
• i.e., generalised velocity:
Assumption
• All generalised velocities are independent of each other.
• i.e.,
 Cont.
Further assumptions
• are independent of each other.

• These assumptions hold

while solving problems
 Cont.
Generalised momentum (𝑝𝑖 )
• It is given by: where L is Lagrangian.

• 𝑞𝑖 , 𝑞ሶ 𝑖 , 𝑝𝑖 may or may not have dimension of distance, velocity, and

momentum
• However product of 𝑞𝑖 and 𝑝𝑖 : [𝑞𝑖 ][𝑝𝑖 ] always has dimension of
angular momentum which are
• If 𝑞𝑖 represents a distance,
then 𝑝𝑖 represents component of actual momentum.
 Cont.
• If 𝑞𝑖 = x, then 𝑝𝑖 = 𝑝𝑥 [dimensions of linear momentum]
• If 𝑞𝑖 = ∅ (angle), then 𝑝𝑖 = 𝑝∅ [dimensions of angular momentum]
• [the angle is dimensionless]
• Generalised momentum corresponding to angle is angular
momentum.
• In spherical polar (r, θ, ∅) , polar (r, ∅) and in cylindrical (s, ∅, z), angle
∅ represents rotation about z axis.
• Therefore 𝑃∅ = 𝐿𝑧 [angular momentum about z-axis]
• There is no assumption about independence of generalised
momentum.
Cont.
 Lagrangian Equation of Motion
Deducing the equation
• It is derived using the following two principles:
• (1) D’Alembert’s principle [uses concept of virtual work]
• (2) Variational principle [uses Euler mathematical method]
• We let T be the kinetic energy of a system, and 𝑄𝑖 its generalised
force, then Euler-Lagrange equation is written as:
 Cont.
• If 𝑄𝑖 can be expressed in terms of some generalised potential
U (𝑞𝑖 , 𝑞ሶ 𝑖 , t) as:

• (monogenic system)

• Equating the T equation to the U equation, then the Euler-Lagrangean

equation of motion becomes:

where L = T - U
 Cont.
• If U depends only on coordinates, then U(𝑞𝑖 ) = V (by notation)

• Then:

• Where is used to find generalised

force when generalised potential is given.

 Worked Example 9
• The Lagrangian of a system is:

Find its equation of motion.

Solution
• The generalised coordinate is 𝑞𝑖 = x. The E-L equation of motion is
given by:
 Cont.
• Replacing the generalised coordinate: 𝑞𝑖 = x:
: = =

= then

⇒
 Worked Example 10

• If

Which type of motion does the above Lagrangian represent?

Solution
• We find the equation of motion first:
• This can be rewritten as:

• Thus: mass x acceleration = Force [the Force is velocity dependent]

• This is damped motion.
 Worked Example 11

Examples with one degree of freedom

Simple pendulum

The pendulum is a one degree of freedom system, with coordinate 𝜃.

The velocity of the mass is 𝑣𝜃 = l𝜃, and distance below the pivot is l
cos(𝜃). The Lagrangian is thus:
 Cont.

The generalized momentum for 𝜃 works out to be mass’s angular

momentum around the pivot:

and the generalized force for 𝜃 works out as the gravitational torque
around the pivot:
 Cont.
Putting these together, we get the correct equation for the pendulum:

We have ended up doing rotational dynamics, with angular momentum

and torque, culminating not in 𝑝ሶ = F but its angular equivalent 𝐽 ሶ = G.
 Lagrangian mechanics problem set
1. Sliding ladder

A ladder (see diagram above) slides without friction with kinetic energy
as the sum of rotational and translational parts as given below:
 Cont.

and the potential energy as:

(i) Write the Lagrangian function of the system.

(ii) The generalised momentum
(iii) The generalised force
(iv) The equation of motion
 Cont.
2. A bead on a wire is described by the arc-length coordinate s, so its
speed is 𝑠ሶ and its height is h(s), write:

(i) the Lagrangian function of the bead

(ii) the generalised momentum
(iii) the generalised force
(iv) the equation of motion of the bead
 Cont.
3. Consider a pendulum made of a spring with a mass m on the end
(see diagram below). The spring is arranged to lie in a straight line
(which we can arrange by, say, wrapping the spring around a rigid
massless rod). The equilibrium length of the spring is ℓ. Let the spring
have length ℓ + x(t), and let its angle with the vertical be θ(t). Assuming
that the motion takes place in a vertical plane, find the equations of
motion for x and θ.