Ch.

6
Manipulator Dynamics

Prof. Dr. Ahmed Badawy

Introduction

Lagrangian mechanics is a mathematical formalism used to

describe the dynamics of physical systems. It was
developed by the French mathematician Joseph-Louis
Lagrange in the late 18th century.
The central idea of Lagrangian mechanics is to describe the
motion of a system using a function called the Lagrangian,
which is defined as the difference between the kinetic
energy and potential energy of the system.
The Lagrangian provides a concise and elegant way to
express the equations of motion for a system, which makes
it an important tool in physics and engineering.
Introduction

There are different methods to derive the dynamic

equations of a dynamic system. As final result, all of them
provide sets of equivalent equations, but their
mathematical description differs with respect to their
eligibility for computation and their ability to give insights
into the original mechanical problem.
Generalized Coordinates

The generalized coordinate values must uniquely define

any possible position of the system relative to the initial
position.

For example, it should be possible to draw a diagram of the

system in its current position by knowing only values of the
generalized coordinates and the fixed dimensions.

The minimum number of generalized coordinates required

to specify the position of the system is the number of
degrees of freedom of that system.
Generalized Coordinates

The linkage in Fig. has one degree of freedom because we are

able to locate the current configuration of the linkage solely from
knowledge of θ. The choice of generalized coordinates is not
unique. Thus, different individuals analyzing the same system
might select different generalized coordinates, and therefore
derive different equations of motion.
Generalized Coordinates

When a rigid body moves in space without any restrictions,

its current configuration is defined by the position of any
point, such as the center of mass, and a set of three
independent direction angles locating lines in the body.
Thus, in the absence of kinematical constraints, any rigid
body moving in space has six degrees of freedom, and a
possible set of generalized coordinates are three position
coordinates of the center of mass relative to a convenient
fixed reference frame and three Euler angles defined relative
to that reference frame.
Generalized Coordinates

Now suppose a body is constrained to execute a planar

motion. This restraint reduces the number of degrees of
freedom to three, because only two position coordinates are
required to locate a point in the plane of motion, and the
only angle of rotation is about the axis perpendicular to that
plane.
 Generalized Coordinates

Consider the rigid bar in Fig. where position coordinates xA and yA

locate one end of the bar, and angle θ measured from the horizontal
locates the current orientation. Only geometrical parameters that
change as the system moves are candidates for generalized
coordinates, whereas fixed parameters, such as L, are considered to
be known system properties. Thus the generalized coordinates
defined in Fig. are (xA, yA, θ)., or (xB, yB, φ), or (xA, yA, xB).
Constraint Equations

Suppose we select a set of N constrained generalized

coordinates to represent a system. We reserve the symbol qj
to denote the jth generalized coordinate. There might be
several constraint equations that these generalized
coordinates must satisfy.
f j  q1 , q2 ,..., qN   0
This relation is referred to as a configuration (scleronomic
or fixed) constraint equation. This term stems from the fact
that, by limiting the values the generalized coordinates
may have, it restricts the overall arrangement of the
system at any instant.
Constraint Equations

xB = xA + L cos(θ),
yB = yA + L sin(θ),
φ = θ + π/2

A constraint is classed as scleronomic if the time t does not

appear explicitly in the equation of constraint. Otherwise, it is
rheonomic.
 Constraint Equations

Suppose that the configuration of a system is specified by N

generalized coordinates (q1,…, qN) and assume that there are
m independent equations of constraint of the form

rheonomic
Constraint Equations

In Lagrangian mechanics, a system is said to be non-

holonomic if its motion is subject to constraints that cannot
be expressed in terms of the generalized coordinates and
their time derivatives. Non-holonomic systems are more
complex than holonomic systems, which are subject only to
constraints that can be expressed in terms of the generalized
coordinates and their time derivatives.
Constraint Equations

Non-holonomic constraints are often expressed as inequalities,

which means that they limit the range of possible motions of
the system. For example, a wheeled vehicle can only move in
certain directions, which makes it a non-holonomic system.

The presence of non-holonomic constraints makes it more

difficult to describe the motion of the system, as the
constraints cannot be incorporated into the Lagrangian in a
straightforward manner. Instead, special techniques, such as
the Lagrange multiplier method or the Hamiltonian-Jacobi
theory, must be used to account for the non-holonomic
constraints and derive the equations of motion for the system.
Equations of motion

d  T  T V
     Qj
dt  q j  q j q j

Where
T ... Kinetic Energy
V ... Potential Energy
q ... Generalized coordinate
Q ... Generalized force
Example 1

Formulate the dynamic equations

for planner RR manipulator

Kinetic Energy
T  T1  T2

T1  0.5 I1,o112
11 2 2
  m1l1  1
23 
 T2  m2  x   
2
1
2
2
G2 y 2
G2
1
2 I 2,G 2 1  2

 m   l  l  l l c     l     
1
2 2 1
2 2
1
1 2
4 2 1 2 2
2
2
1 2
4 2 1 2 l  l1l2 c2 
1 2
2 2

+  m l     
2
1 1 2
2 12 2 2 1 2

11 
T   m1l12 12  12 m2 12  l12  14 l22  l1l2 c2   22  14 l22   12  12 l22  l1l2 c2 
23 
   
2
+ 1
2
1
12 ml 2
2 2 1  2
T    m l  m l  m l  12 m2l1l2 c2  
1
2 1
6
2
11
1
2
2
2 1
1
6
2
2 2

22  16 m2l22   12  13 m2l22  12 m2l1l2 c2 

T
 1  3 m1l1  m2 l1  3 m2 l2  m2l1l2 c2  
 1 2 2 1 2

1
 2  3 m2 l2  12 m2l1l2 c2 
 1 2

d  T 
   
 1 3 1 1
1
m l 2
 m l 2
 3 2 2  m2 l1l2 c2   1 2 m2 l1l2 s2
1
m l 2

dt  1 
2 1

 2  13 m2 l22  12 m2 l1l2 c2   12 22 m2 l1l2 s2

T    m l  m l  m l  12 m2l1l2 c2  
1
2 1
6
2
11
1
2
2
2 1
1
6
2
2 2

22  16 m2l22   12  13 m2l22  12 m2l1l2 c2 

T
  2  3 m2 l2   1  3 m2 l2  12 m2 l1l2 c2 
 1 2
  1 2

 2

d  T 
    
2 3
1
m l 2
2 2    
1 3
1
m l 2
 1

2 m2 l1l2 c2 
dt   2 
2 2

 1 2   12 m2 l1l2 s2 
T    m l  m l  m l  12 m2l1l2 c2  
1
2 1
6
2
11
1
2
2
2 1
1
6
2
2 2

22  16 m2l22   12  13 m2l22  12 m2l1l2 c2 

T
 zero
1

T
  2 m2 l1l2 s21  12 m2 l1l2 s21 2
1 2

 2
Potential Energy
V  0.5m1 gl1s1  m2 g  l1s1  0.5l2 s12 

V 1
 2 m1 gl1c1  m2 g  l1c1  12 l2 c12 
1

V 1
 2 m2 gl2 c12
 2
 d  T  T V
     Qj
dt  q j  q j q j

Lagrange Equation (1)

 1  1  13 m1l12  m2l12  13 m2l22  m2l1l2 c2   12 m2 l1l2 s2
  2  13 m2 l22  12 m2 l1l2 c2   12 22 m2 l1l2 s2
 12 m1 gl1c1  m2 g  l1c1  12 l2 c12 
Lagrange Equation (2)
 2   2  13 m2 l22   1  13 m2 l22  12 m2 l1l2 c2 
 12 m2 l1l2 s212  12 m2 gl2 c12
 Example 2

Formulate the dynamic equations

for planner RP manipulator
Kinetic Energy
T 1
2  1
3 m1l12  2  12 m2 vc22  12  121 m2 l22  2

v d d 
2
c2
2 2 2

T  1
2  1
3  
m1l12  2  12 m2 d 2  d 2 2  12  121 m2l22  2

T   16 m1l12  12 m2 d 2  241 m2l22  2  12 m2 d 2

Potential Energy
V  12 m1 gl1s  m2 gds

V 1
 2 m1 gl1c  m2 gdc

V
 m2 gs
d
T   16 m1l12  12 m2 d 2  241 m2l22  2  12 m2 d 2

T
  3 m1l1  m2 d  12 m2l2 
1 2 2 1 2


d  T  1 2
 3 1 1
   2 2     2m2 d  d 
2 2

1
m l m d m l
dt   
2 12

T T
 m2 d  zero
d 
d  T  T
   m d  m d  2
dt  d 
2
d
2
 d  T  T V
     Qj
dt  q j  q j q j

Lagrange Equation (1)

 1   13 m1l12  m2 d 2  121 m2 l22    2m2 d  d
 12 m1 gl1c  m2 gdc

Lagrange Equation (2)

F2  m2 d  m2 d 2  m2 gs
State Space

state space describes the complete internal state of the

robot, including its position, velocity, acceleration, and any
other relevant variables.

The state space is an important concept in robotics and

control because it allows us to represent the
complete dynamic behavior of the robot in a concise and
abstract way. By analyzing the structure and properties of
the state space, we can design controllers and algorithms
that can manipulate the internal state of the robot to
achieve a desired behavior or task.
State Space

For example, in feedback control, the state space can be

used to design controllers that regulate the internal state of
the robot to achieve a desired state or trajectory. By
measuring the state variables of the robot and comparing
them to the desired state, the controller can compute the
necessary inputs, such as motor torques or control signals,
to control the robot.
 State Space Equation

τ = M  θ  θ + V  θ, θ  + G  θ 
Where
M  θ  ... MassMatrix(n×n)
V  θ, θ  ... Inertia &Coriolis vector (n  1)
G  θ  ... Gravity vector (n  1)

We use sate space equation as V has both position and

velocity dependence
Configuration Space

Configuration space is a mathematical construct that is

used to represent the set of all possible configurations of a
robot. A configuration is a set of values that describe the
position and orientation of the robot's joints or end-effector.
Configuration Space

Assume we have a set of position vectors define the

configuration of the system in real (physical) space. If we
choose (q1,…, qN) as generalized coordinates of RN, then
each set of variables (q1,…, qN) will represent a point in the
configuration space. So, any configuration of the system is
represented by one point in the configuration space RN.

As the mechanical system changes its configuration with

time, the configuration point traces a curve in configuration
space called generalized trajectory.
Configuration Space

In a 2D planar robot with two revolute joints (RR robot),

the configuration of the robot can be described by two
angles, q1 and q2, that represent the orientations of the
two joints. The configuration space of the robot is
therefore a two-dimensional space, where each point in
the space corresponds to a unique configuration of the
robot.
Configuration Space Equation

If we split V into two parts B & C

τ = M  θ  θ + B  θ  θθ  + C  θ  θ 2  + G  θ 

B  θ Matrix of order n x n(n-1)/2 of Coriolis coefficients

 θθ  Vector n(n-1) x 1 of joint velocity product

C  θ Matrix n x n for inertia terms

Importance of configuration space that all matrices contain

position only, so more suitable for computer control
 Example 3

Rewrite the dynamic equation of example 1 in state space

equation form
 13 m1l12  m2 l12  13 m2 l22  m2 l1l2 c2 1
m2 l22  12 m2 l1l2 c2 
M θ   3


1
3 m l
2 2
2
 1
2 m2 l1l2 c2
1
3 m l
2 2
2

  m2 l1l2 s21 2  12 m2 l1l2 s2 22 
V  θ, θ    
  2 m2 l1l2 s21
1 2

 12 m1 gl1c1  m2 g  l1c1  12 l2 c12  
G θ   
 
1
2 m2 gl c
2 12
 Example 4

Rewrite the dynamic equation of example 1 in configuration

space equation form

 m2 l1l2 s2   0  12 m2 l1l2 s2 

B θ    C θ   1 
 0   2 m2 l1l2 s2 0 

12 
τ = M  θ  θ + B  θ  1 2  + C  θ   2  + G  θ 
 2 