Can be found at: http://www.cs.unc.edu/~kim/matlab.

ppt

Intro to Matlab

1. Using scalar variables

2. Vectors, matrices, and arithmetic
3. Plotting
4. Solving Systems of Equations
 New Class--for Engineers
• ENGR196.3
• SPECIAL TOPICS: INTRODUCTION TO
MATLAB
• Description: Fundamentals of MATLAB
programming applied to problems in science and
mathematics. Solving systems of equations, basic
scripting, functions, vectors, data files, and graphics.
(Credit course for grade or CR/NC)
 Why use Matlab?
• Advantages:
Handles vector and matrices very nice
Quick plotting and analysis
EXTENSIVE documentation (type ‘help’)
Lots of nice functions: FFT, fuzzy logic, neural
nets, numerical integration, OpenGL (!?)
One of the major tools accelerating tech change
• Drawbacks:
Slow (execution) compared to C or Java
 A tour of Matlab’s features
• Click on Help>Full Product Family Help:
Check out Fuzzy Logic
Genetic Algorithms
Symbolic Math
 Scalars
• The First Time You Bring Up MATLAB
• MATLAB as a Calculator for Scalars
• Fetching and Setting Scalar Variables
• MATLAB Built-in Functions, Operators,
and Expressions
• Problem Sets for Scalars
 3-1 The First Time You Bring Up
MATLAB
Basic windows in MATLAB are:
• Command - executes single-line commands
• Workspace - keeps track of all defined variables
• Command History - keeps a running record of all single
line programs you have executed
• Current Folder - lists all files that are directly available for
MATLAB use
• Array Editor - allows direct editing of MATLAB arrays
• Preferences - for setting preferences for the display of
results, fonts used, and many other aspects of how
MATLAB looks to you
 3-2 MATLAB as a Calculator for
Scalars
• A scalar is simply a number…

22
3.5, ,  4, 
7
• In science the term scalar is used as opposed to a
vector, i.e. a magnitude having no direction.
• In MATLAB, scalar is used as opposed to arrays, i.e. a
single number.
• Since we have not covered arrays (tables of numbers)
yet, we will be dealing with scalars in MATLAB.
 3-3 Fetching and Setting Scalar
Variables
• Think of computer
variables as named
containers.
• We can perform 2
types of operations
on variables:

we can set the value held in the container: x = 22

we can look at the value held in the container: x
 The Assignment Operator (=)
• The equal sign is the assignment operator
in MATLAB.
>> x = 22
places number 22 in container x
• How about:
>> x = x + 1
• Note the difference between the equal sign
in mathematics and the assignment
operator in MATLAB!
 Useful Constants
• Inf infinity
• NaN Not a number (div by zero)
• eps machine epsilon
• ans most recent unassigned answer
• pi 3.14159….
• i and j Matlab supports imaginary
numbers!
Using the Command History
Window
 3-4 MATLAB Built-in Functions,
Operators, and Expressions
• MATLAB comes with a large number of built-in
functions (e.g.. sin, cos, tan, log10, log, exp)
• A special subclass of often-used MATLAB
functions is called operators
– Assignment operator (=)
– Arithmetic operators (+, -, *, /, ^)
– Relational operators (<, <=, = =, ~=, >=, >)
– Logical operators (&, |, ~)
Example – Arithmetic Operators

Hint: the function exp(x) gives e raised to a power x

Example – Relational and Logical
Operators
Vector Operations

Chapter 5
 Vector Operations
• Vector Creation
• Accessing Vector Elements
• Row Vectors and Column Vectors, and the
Transpose Operator
• Vector Built-in Functions, Operators, and
Expressions
 Vectors and Matrices
• Can be to command line or from *.m file

scalar: x=3
vector: x = [1 0 0]
2D matrix: x = [1 0 0; 0 1 0; 0 0 1]
arbitrarily higher dimensions (don’t use much)

• Can also use matrices / vectors as elements:

x = [1 2 3]
y = [ x 4 5 6]
2-D Plotting and Help in
MATLAB
Chapter 6
 2-D Plotting and Help in MATLAB

• Using Vectors to Plot Numerical Data

• Other 2-D plot types in MATLAB
• Problem Sets for 2-D Plotting
6-2 Using Vectors to Plot Numerical Data
• Mostly from observed data - your goal is to
understand the relationship between the variables
of a system.
Speed (mi/hr) 20 30 40 50 60 70

Stopping Distance (ft) 46 75 128 201 292 385

• Determine the independent and dependent

variables and plot:
speed = 20:10:70;
stopDis = [46,75,128,201,292,385];
plot(speed, stopDis, '-ro') % note the ‘-ro’ switch
• Don’t forget to properly label your graphs:
title('Stopping Distance versus Vehicle Speed', 'FontSize', 14)
xlabel('vehicle speed (mi/hr)', 'FontSize', 12)
ylabel('stopping distance (ft)', 'FontSize', 12)
grid on
Sample Problem – Plotting
Numerical Data
 3D Plotting
• 3D plots – plot an outer product

x = 1:10
y = 1:10
z = x’ * y
mesh(x,y,z)

Single quote ‘ means transpose (can’t cut

and paste though…)
 Flow Constructs
• IF block • WHILE block
if (<condition>) while (<condition>)
<body> <body>
elseif end
<body>
end
Conditions same as C, ( ==, >=, <=) except != is ~=
 More Flow Constructs
• FOR block • SWITCH statement

for i = 1:10 switch <expression>

<body> case <condition>,
end <statement>
otherwise <condition>,
<statement>
end
 Other Language Features
• Matlab language is pretty sophisticated
– Functions
Stored in a *.m file of the same name:
function <return variable> = <function name>(<args>)
<function body>
– Structs
• point.x = 2; point.y = 3; point.z = 4;
– Even has try and catch (never used them)
 Solving Systems of Equations
• Consider a system of simultaneous equations
3x + 4y + 5z = 32
21x + 5y + 2z = 20
x – 2y + 10z = 120
• A solution is a value of x, y, and z that satisfies
all 3 equations
• In general, these 3 equations could have
1 solution, many solutions, or NO solutions
 Using Matlab to Solve
Simultaneous Equations
• Set up the equation in matrix/vector form:

3 4 5 x 32
21 5 2 * y = 20
1 -2 10 z 120

A = [3 4 5; 21 5 2; 1 -2 10]
u = [ x y z]’
b = [ 32 20 120]’
In other words, A u = b (this is linear algebra)
 The solution uses matrix inverse
• If you multiply both sides by 1/A you get
u = 1/A * b

• In the case of matrices, order of operation is

critical (WRONG: u = b/A )

• SO we have “Left division” u=A\b

(recommended approach)
• OR use inv( ) function: u = inv(A) * b
 The solution
>> u = A\b
u=
1.4497 ( value of x)
-6.3249 ( value of y)
10.5901 ( value of z)

• You can plug these values in the original

equation test = A * u and see if you get b
 Caution with Systems of Eqs
• Sometimes, Matrix A does not have an
inverse:
1 -1 0 x 32
1 0 -1 * y = 10
z 22
0 1 -1
• This means the 3 equations are not really
independent and there is no single solution
(there may be an infinite # of solns)
• Take determinant det(A) if 0, it’s singular