Control Example using Matlab

Cruise Control

Modeling a Cruise Control System

Physical setup and system equations

The inertia of the wheels is

neglected
Aerodynamic Drag is neglected
is proportional to the square
of the cars speed
It is assumed that friction is opposing the motion of the car
is proportional to the car's speed
The problem is reduced to the simple mass and damper system

Modeling a Cruise Control System

Using Newton's law, the dynamic equation
for this system is:

mxbx u

where u is the force from the engine.

There are several different ways to describe a system of
linear differential equations.
To calculate the Transfer Function, we shall use the Stare
Space representation then transform it to TF using ss2tf

State-space equations
The state-space representation is given by the
equations:

dx
A x Bu
dt
y C x Du

where

x is an n by 1 vector representing the state

(commonly position and velocity variables in mechanical systems),

u is a scalar representing the input
(commonly a force or torque in mechanical systems), and
y is a scalar representing the output.

State Equation for our CC model

The state vector x is [ x, x ] and y is x

The equation

mx bx u

is

b
1

u
x
x
m
m

or

1 x 0
0
0 b 1 u
x
m x m
x
y 0 1 0 u
x
x

Design requirements
For this example, let's assume that
m = 1000kg
b = 50Nsec/m
u = 500N
Design requirements
When the engine gives a 500 Newton force, the car will reach a
maximum velocity of 10 m/s (22 mph).
An automobile should be able to accelerate up to that speed in less
than 5 seconds. Since this is only a cruise control system, a 10%
overshoot on the velocity will not do much damage. A 2% steadystate error is also acceptable for the same reason.
Rise time < 5 sec
Overshoot < 10%
Steady state error < 2%

Matlab representation and open-loop response

m = 1000;
b = 50;
u = 500;
A = [0 1; 0 -b/m]
B = [0; 1/m]
C = [0 1]
D=0
step(A,u*B,C,D)

this step response does not meet the design criteria placed on the
problem. The system is overdamped, so the overshoot response is
fine, but the rise time is too slow.

P controller
The first thing to do in this problem is to transform the state-space
equations into transfer function form.
m = 1000; b = 50; u = 500; A = [0 1; 0 -b/m] B = [0; 1/m] C = [0 1] D = 0

[num,den]=ss2tf(A,B,C,D)
Next, close the loop and add some proportional control to see if the
response can be improved.

k = 100;
[numc,denc] = cloop(k*num,den,-1);
function

%Closed-loop transfer

t = 0:0.1:20;
step(10*numc,denc,t) %for 10m/sec the signal used by
step
axis([0 20 0 10])

Get the Transfer Function

For
m = 1000; b = 50; u = 500; A = [0 1; 0 -b/m] B = [0; 1/m] C = [0 1] D
=0

[num,den]=ss2tf(A,B,C,D)
Matlab should return the following to the command window:
num =
0

0.0010

den =
1.0000 0.0500

This is the way Matlab presents the TF

0.001s + 0
-----------------s^2 + 0.05s + 0

Step function
The step function is one of most useful functions in
Matlab for control design.
Given a system that can be described by either a
transfer function or a set of state-space equations,
the response to a step input can immediately be
plotted.
A step input can be described as a change in the
input from zero to a finite value at time t = 0.
By default, the step command performs a unit step
(i.e. the input goes from zero to one at time t = 0).
The basic command to use the step function is one
of the following
(depending if you have a set of state-space equations or a
transfer function form):
step(A,B,C,D)
step(num,den)

P controller
The steady state error is more than 10%,
and the rise time is still too slow.
Adjust the proportional gain to make the
response better.
but you will not be able to make
the steady state value go to 10m/sec
without getting rise times that are too
fast.

You must always keep in mind that you are designing a real system,
and for a cruise control system to respond 0-10m/sec in less than half
of a second is unrealistic.
To illustrate this idea, adjust the k value to equal 10,000, and you
should get the following velocity response:

P controller
The solution to this problem is to add some
integral control to eliminate the steady state
error.
Adjust k until you get a reasonable rise time.
For example, with k = 600, the response should now
look like:

PI controller
Remember, integral control makes the transient response worse, so adding too
much initially can make the response unrecognizable.

k = 600;
ki = 1; % small to get feeling for integral response
num1 = [k ki]; % PI has two gains
den1 = [1 0];
num2 = conv(num,num1); % convolution used to
multiply the
% polynomials representing the gains
den2 = conv(den,den1);
[numc,denc] = cloop(num2,den2,-1);
t = 0:0.1:20;
step(10*numc,denc,t)
axis([0 20 0 10])

PI controller
Now you can adjust
the ki value to reduce
the steady state error.
With ki = 40 and k
adjusted up a little
more to 800, the
response looks like
the following:
As you can see, this step
response meets all of the
design criteria, and
therefore no more iteration
is needed.
It is also noteworthy that
no derivative control was
needed in this example

Modeling a Cruise Control System

Physical setup and system equations

Lets now add Drag and assume the friction to be non velocity
dependent (the static friction, proportional to normal force).
dv(t )
M
Cu (t ) Bv (t )
dt
2

- M
- B
- C

d (t )
(t ) u (t )
dt
2

:
:

max

max

C
B

CB
M

0 u (t ) 1

 Simulink

]C = 6250 [N
]B = 2.5 [N/(m/s)2
]M = 1250 [Kg

: 0 ,1 - -
Saturation

Very Short Simulink Tutorial

In the Matlab command window write simulink.
The window that has opened is the Simulink Library Browser.
It is used to choose various Simulink modules to use in your simulation.

From this window, choose the File menu, and then New
(Model).
Now we have a blank window, in which we will build our model.
This blank window and the library browser window, will be the
windows well work with.

We choose components from the library browser, and then drag

them to our work window.
Well use only the Simulink library (also called toolbox) for now.

Simulink Tutorial
As we can see, the Simulink library is divided into several
categories:

1.

Continuous Provides functions for continuous

time, such as integration, derivative, etc.

2.

Discrete Provides functions for discrete time.

3.

Funcitons & Tables Just what the name says.

4.

Math Simple math functions.

5.

Nonlinear Several non-linear functions, such as

switches, limiters, etc.

6.

Signals & Systems Components that work

with signals. Pay attention to the mux/demux.

7.

Sinks Components that handle the outputs of the

system (e.g. display it on the screen).

8.

Sources Components that generate source signals

for the system.

Simulink Tutorial: First simulation

Drag the Constant component

from the Simulink library
(Sources) to the work window,
and then drag the Scope module
(Sinks) in the same manner.
Now, click the little triangle
(output port) to the right of the
Constant module, and while
holding the mouse button down,
drag the mouse to the left side of
the scope (input port) and then
release it. You should see a
pointed arrow being drawn.

Double click the Constant module to open its dialog window.

Now you can change this modules parameters.
Change the constant value to 5.

Simulink Tutorial: First simulation

Double-click the scope to view its window.

You can choose Simulation

Parameters from the Simulation to
change the time limits for your simulation.
Choose Start from the Simulation menu
(or press Ctrl+T, or click the play button
on the toolbar) to start the simulation.

Simulink Tutorial: First simulation

Choose Start from the Simulation menu
(or press Ctrl+T, or click the play button
on the toolbar) to start the simulation.
Now this is a rather silly simulation. All
it does is output the constant value 5 to the
graph (the x-axis represents the simulation
time).

Right-click on the scopes graph window

and choose Autoscale to get the following
result:

Simulink Tutorial
Now lets try something a little
bit more complicated.
First, build the following
system (you can find the Clock
module in the Sinks category.
The Trigonometric Function
and Sum modules reside in the
Math category):

Simulink Tutorial
Now, lets see if the
derivative is really a cosine.
Build the following system
(the Derivative module is
located in the Continuous
category):

We can see that this is indeed a

cosine, but something is wrong
at the beginning.

Simulink Tutorial
This is because at time 0,
the derivative has no
prior information for
calculation
theres no initial value for
the derivative,
so at the first time step,
the derivative assumes
that its input has a
constant value (and so the
derivative is 0).

An other example
Lets say that we have a differential equation that we want
to model. The equation is:

A 0.5 A

A 0.5
0

How can we solve this numerically using Simulink?

Well notice 2 simple facts:
1. If we have A, then we have A' (multiplication).
2. If we have A', then we have A (integration).
We can get out of this loop by using the initial condition. We know that
A0 = 0.5, so now we can calculate A' and then recalculate the new
A, and so on.

Simulink Tutorial
double-click the
Integrator and choose
Initial Condition Source:
External.
Note that pressing ctrl
when clicking the mouse
button on a line, allows
you to split it into 2 lines:

Set the simulation stop time to

3.5 seconds
the solution goes to infinity

and see the results in the scope:

Simulink
Automatic Cruise Control

C = 6250 [N]
B = 2.5 [N/(m/s)2]
M = 1250 [Kg]

- ,1 - 0 :
Saturation

Automatic Cruise Control

(plant)

Automatic Cruise Control

Automatic Cruise Control

(0) 0.5

0
.8

,P I ) PI (.
: :
. ) (t "" .
- "" ) (.

Automatic Cruise Control

P controller No wind

Automatic Cruise Control

P controller with wind

Automatic Cruise Control

I controller No wind

Automatic Cruise Control

PI controller No wind

Automatic Cruise Control

PI controller with wind

Automatic Cruise Control

Automatic Cruise Control