Linear, Time-Invariant (LTI)

Longitudinal Model

Time Response
Robert Stengel, Aircraft Flight Dynamics
MAE 331, 2010

Steady, level flight

Simplified control effects
Neglect disturbance effects

Time-domain analysis

$
&
&
&
&
&
%

Transient response to initial conditions and inputs

Steady-state (equilibrium) response
Continuous- and discrete-time models
Phase-plane plots
Response to sinusoidal input

$ *DV
&
!V! (t) ' & L
)
V
VN
!"! (t) ) &
)=& M
!
!q(t)
V
) &
!#! (t) )( & LV
& *
VN
&%

*g
0

*Dq

*D#

Lq

L#

VN

Mq

VN

M#
*

L#

VN

'
)$
)&
)&
)&
)&
)&
)%
)(

!V (t) ' $ 0
) &
!" (t) ) & 0
+
!q(t) ) & M + E
) &
!# (t) )( & 0
%

T+ T

L+ F / VN

*L+ F / VN

'
) $ !+ E(t) '
)
)&
) & !+ T (t) )
) & !+ F(t) )
(
)%
(

What can we do with it?

Copyright 2010 by Robert Stengel. All rights reserved. For educational use only.
http://www.princeton.edu/~stengel/MAE331.html
http://www.princeton.edu/~stengel/FlightDynamics.html

Integrate equations to obtain time histories of initial condition, control, and

disturbance effects
Determine modes of motion
Gain insights about
Examine steady-state conditions
system dynamics
Identify effects of parameter variations
Define frequency response

System Response to Inputs

and Initial Conditions

Linear, Time-Invariant
System Model

General model contains

Solution of the linear, time-invariant (LTI) dynamic model

!!x(t) = F!x(t) + G!u(t) + L!w(t), !x(t o ) given

Dynamic equation (ordinary differential equation)

Output equation (algebraic transformation)

!x(t) = !x(t o ) + # [ F!x(" ) + G!u(" ) + L!w(" )]d"

to

!x (t) = F!x(t) + G!u(t) + L!w(t), !x(t o ) given

!y(t) = H x !x(t) + H u !u(t) + H w !w(t)
State and output dimensions need not be the same

dim [ !x(t)] = ( n " 1)

dim [ !y(t)] = ( r " 1)

... has two parts

Unforced (homogeneous) response to initial conditions
Forced response to control and disturbance inputs

Unforced Response
to Initial Conditions
Neglecting forcing functions

Response to
Initial Conditions

!x(t) = !x(t o ) + # [ F!x(" )]d" = eF (t $to ) !x(t o ) = % ( t $ t o ) !x(t o )

to

The state transition matrix, !, propagates the

state from to to t by a single multiplication
eF (t !to ) = Matrix Exponential
2
3
1
1
"# F ( t ! t o ) $% + "# F ( t ! t o ) $% + ...
2!
3!
= & ( t ! t o ) = State Transition Matrix

= I + F (t ! to ) +

Initial-Condition Response
via State Transition
Propagation of "x
!x(t1 ) = " ( t1 # t o ) !x(t o )

!x(t 2 ) = " ( t 2 # t1 ) !x(t1 )

Response to continuous controls and disturbances

!x(t k +1 ) = !x(t k ) +

!x(t1 ) = " (# t ) !x(t o ) = "!x(t o )

!x(t 2 ) = "!x(t1 ) = " 2 !x(t o )
!x(t 3 ) = "!x(t 2 ) = " 3 !x(t o )

1
1
2
3
! = I + F (" t ) + #$ F (" t ) %& + #$ F (" t ) %& + ...
2!
3!

t k+1

!x(t 3 ) = " ( t 3 # t 2 ) !x(t 2 )

If (tk+1 tk) = !t = constant,

state transition matrix is
constant

Discrete-Time Dynamic Model

# [ F!x(" ) + G!u(" ) + L!w(" )]d"

tk

Response to piecewise-constant controls and disturbances

!x(t k +1 ) = " (# t ) !x(t k ) + " (# t )

t k+1

* &' e

$ F (% $t k )

tk

= "!x(t k ) + +!u(t k ) + ,!w(t k )

! = eF" t

(
% = (e

)
$ I) F

# = eF" t $ I F $1G
F" t

$1

(d% [ G!u(t k ) + L!w(t k )]

)
Ordinary Difference Equation

Control- and Disturbance-Effect

Matrices

Discrete-Time Response to Inputs

!x(t k ) = "!x(t k #1 ) + $!u(t k #1 ) + %!w(t k #1 )

!x(t1 ) = "!x(t o ) + #!u(t o ) + $!w(t o )

!x(t 2 ) = "!x(t1 ) + #!u(t1 ) + $!w(t1 )

Propagation of "x, with constant !, #, and $

! = eF" t # I F #1G

!x(t 3 ) = "!x(t 2 ) + #!u(t 2 ) + $!w(t 2 )

!

1
1
1
$
'
= & I # F" t + F 2" t 2 # F 3" t 3 + ...) G" t
%
(
2!
3!
4!

* = eF" t # I F #1L
1
1
1
$
'
= & I # F" t + F 2" t 2 # F 3" t 3 + ...) L" t
%
(
2!
3!
4!

! t = t k +1 " t k

As !t becomes very small

! $"$$
# ( I + F" t )
t #0
% $"$$
# G" t
t #0
# L" t
& $"$$
t #0

Continuous- and Discrete-Time

Short-Period System Matrices

Continuous-time
(analog) system

! t = t k +1 " t k

!t = 0.01 s

# 0.987
!=%
$ 0.01

# "0.079 &
*=%
(
$ "0.013 '

!t = 0.1 s

# 0.845
!=%
$ 0.0869

"0.694 &
(
0.846 '

# "0.84 &
)=%
(
$ "0.0414 '

" !7.9856 %
L=$
'
# !1.2709 &

# "0.694 &
*=%
(
$ "0.154 '

!t = 0.5 s
!t has a large effect
on the digital model

# 0.0823
!=%
$ 0.185

Learjet 23
MN = 0.3, hN = 3,050 m
VN = 98.4 m/s

"0.079 &
(
0.987 '

# "0.09 &
)=%
(
$ "0.0004 '

" !1.2794 !7.9856 %

F=$
'
1
!1.2709 &
#
" !9.069 %
G=$
'
0
#
&

Discrete-time (digital) system

Example: Aerodynamic
Angle, Linear Velocity, and
Angular Rate Perturbations

"1.475 &
(
0.0839 '

# "2.492 &
)=%
(
$ "0.643 '
# "1.475 &
*=%
(
$ "0.916 '

Aerodynamic angle and linear velocity perturbations

!" ! !w V ; !" = 1 # !w = 0.01745 $ 98.4 = 1.7 m s

N

!% ! !v V ; !% = 1 # !v = 0.01745 $ 98.4 = 1.7 m s

N

Angular rate and linear velocity perturbations

!p = 1 / s; !wwingtip = !p "# b 2 $% = 0.01745 & 5.25 = 0.09 m s

!q = 1 / s; !wnose = !q [ xnose ' xcm ] = 0.01745 & 6.4 = 0.11m s

!r = 1 / s; !vnose = !r [ xnose ' xcm ] = 0.01745 & 6.4 = 0.11m s

Example: Continuous- and

Discrete-Time Models
Learjet 23
MN = 0.3, hN = 3,050 m
VN = 98.4 m/s

Difference Equations
Produce State Increments

Short
Period

# !q! & # )1.3 )8 & # !q & # )9.1 &

(=%
(+%
%
(%
( !* E
)1.3 ' %$ !" (' $ 0 '
%$ !"! (' $ 1

# !qk +1 & # 0.85 )0.7 & # !qk & # )0.84 &

(=%
(+%
%
(%
( !* Ek
%$ !" k +1 (' $ 0.09 0.85 ' %$ !" k (' $ )0.04 '

RollSpiral

# !!p & # *1.2 0 & # !p & # 2.3 &

%
()%
(+%
(%
( !+ A
0 ' % !" ( $ 0 '
%$ !"! (' $ 1
$
'

# !pk +1 & # 0.89 0 & # !pk & # 0.24 &

()%
(+%
%
(%
( !+ Ak
%$ !" k +1 (' $ 0.09 1 ' %$ !" k (' $ *0.01 '

# !!r & # *0.11 1.9 & # !r & # *1.1 &

% ! ()%
(+%
( !+ R
(%
%$ !" (' $ *1 *0.16 ' %$ !" (' $ 0 '

" !!x1 % " (1.2794 (7.9856 % " !x1 % " (9.069 %

'=$
'+$
$
'$
' !) E
1
(1.2709 & $ !x2 ' #
0
$# !!x2 '& #
&
&
#

! t = 0.1sec

Differential Equations Produce

State Rates of Change

Dutch
Roll

Initial-Condition Response

# !rk +1
%
%$ !" k +1

& # 0.98 0.19 & # !rk

()%
(%
(' $ *0.1 0.97 ' %$ !" k

Angle of Attack
Initial Condition

" !y1 % " 1 0 % " !x1 % " 0 %

'=$
'+$
$
'$
' !) E
$# !y2 '& # 0 1 & $# !x2 '& # 0 &

Short-Period Linear Model - Initial Condition

F
G
Hx
sys

=
=
=
=

[-1.2794 -7.9856;1. -1.2709];

[-9.069;0];
[1 0;0 1];
ss(F, G, Hx,0);

xo
[y1,t1,x1]

=
=

[1;0];
initial(sys, xo);

xo
=
[2;0];
[y2,t2,x2]
=
initial(sys, xo);
plot(t1,y1,t2,y2), grid

& # *0.11 &

(+%
( !+ Rk
(' $ 0.01 '

Pitch Rate
Initial Condition

figure
xo
=
[0;1];
initial(sys, xo), grid

Note individual acceleration and difference sensitivities to state and control perturbations

Doubling the initial condition doubles the output

State (Phase)-Plane Plots

" !!x1 % " 0
'($
$
2
$# !!x2 '& $# )* n
%

2nd-Order Model - Initial Condition Response

clear
z
wn
F
G
Hx
sys
t
xo
[y1,t1,x1]

Phase Plane Plots

% " !x1 % " 1 )1 % " !u1 %

1
'+
'
$
'$
)2+* n ' $ !x2 ' $# 0 2 '& $ !u2 '
&#
&
&
#

=
=
=
=
=
=
=
=
=

0.1; % Damping ratio

6.28; % Natural frequency, rad/s
[0 1;-wn^2 -2*z*wn];
[1 -1;0 2];
[1 0;0 1];
ss(F, G, Hx,0);
[0:0.01:10];
[1;0];
initial(sys, xo, t);

plot(t1,y1)
grid on
figure
plot(y1(:,1),y1(:,2))
grid on

Cross-plot of one component against

another
Time or frequency not shown explicitly

Dynamic Stability Changes

the State-Plane Spiral

Damping ratio = 0.1

Damping ratio = 0.3

Damping ratio = 0.1

Superposition of
Linear Responses

Step Response
%

Short-Period Linear Model - Step

F
G
Hx
sys
sys2

=
=
=
=
=

[-1.2794 -7.9856;1. -1.2709];

[-9.069;0];
[1 0;0 1];
ss(F, -G, Hx,0); % (-1)*Step
ss(F, -2*G, Hx,0); % (-1)*Step

Step response
step(sys, sys2), grid

Stability, speed of response,

and damping are
independent of the initial
condition or input

%' 0, t < 0
!" E ( t ) = &
#1, t $ 0
('
" !!x1 % " (1.2794 (7.9856 % " !x1 % " (9.069 %
'=$
'+$
$
'$
' !) E
1
(1.2709 & $ !x2 ' #
0
$# !!x2 '& #
&
&
#
" !y1 % " 1 0 % " !x1 % " 0 %
'=$
'+$
$
'$
' !) E
$# !y2 '& # 0 1 & $# !x2 '& # 0 &

Superposition of Linear Responses

" !!x1 % " (1.2794 (7.9856 % " !x1 % " (9.069 %
'=$
'+$
$
'$
' !) E
1
(1.2709 & $ !x2 ' #
0
$# !!x2 '& #
&
&
#
%

Short-Period Linear Model - Superposition

F
G
Hx
sys

=
=
=
=

[-1.2794 -7.9856;1. -1.2709];

[-9.069;0];
[1 0;0 1];
ss(F, -G, Hx,0); % (-1)*Step

xo
t
u

=
=
=

[1; 0];
[0:0.2:20];
ones(1,length(t));

[y1,t1,x1]
[y2,t2,x2]

=
=

lsim(sys,u,t,xo);
lsim(sys,u,t);

u
[y3,t3,x3]

=
=

zeros(1,length(t));
lsim(sys,u,t,xo);

" !y1 % " 1 0 % " !x1 % " 0 %

'=$
'+$
$
'$
' !) E
$# !y2 '& # 0 1 & $# !x2 '& # 0 &

plot(t1,y1,t2,y2,t3,y3), grid

Doubling the input

doubles the output

Stability, speed of response, and damping are

independent of the initial condition or input

Example: Continuous- and

Discrete-Time LTI
Longitudinal Models

Learjet 23
MN = 0.3, hN = 3,050 m
VN = 98.4 m/s

Learjet 23
MN = 0.3, hN = 3,050 m
VN = 98.4 m/s

Example: Superposition of
Continuous- and Discrete-Time
Longitudinal Models
Phugoid and Short Period

Differential Equations Produce

State Rates of Change

Short
Period

# !q! & # )1.3 )8 & # !q & # )9.1 &

(=%
(+%
%
(%
( !* E
)1.3 ' %$ !" (' $ 0 '
%$ !"! (' $ 1

Phugoid

# !V! & # *0.02 *9.8 & # !V & # 4.7 &

()%
%
(+%
(%
( !+ T
0 ' $% !" (' $ 0 '
%$ !"! (' $ 0.02

Difference Equations
Produce State Increments

# !qk +1
%
%$ !" k +1
# !Vk +1
%
%$ !" k +1

& # 0.85 )0.7 & # !qk

(=%
(%
(' $ 0.09 0.85 ' %$ !" k

& # )0.84 &

(+%
( !* Ek
(' $ )0.04 '

& # 1
)0.98 & # !Vk & # 0.47 &
(=%
(+%
(%
( !* Tk
1
(' $ 0.002
' %$ !" k (' $ 0.0005 '

! t = 0.1sec

$
&
&
&
&
&%

Differential Equations
Produce State Rates of
Change

Difference
Equations Produce
State Increments

$
&
&
&
&
&
%

!V! ' $ *0.02 *9.8

0
0
)
!"! ) & 0.02
0
0
1.3
=&
0
0
*1.3 *8
!q! ) &
) &
0
1
*1.3
!#! )( % *0.002

'$
)&
)&
)&
)&
( &%

$
!Vk +1 ' $
1
*0.98 *0.002 *0.06 ' &
)
)
!" k +1 ) & 0.002
1
0.006 0.12 ) &
&
=
0
0.84 *0.69 ) &
!qk +1 ) & 0.0001
) &
&
0.84 )( &
!# k +1 ) % *0.0002 0.0001 0.09
(
%

!V ' $ 4.7
0 '
)
)
!" ) & 0
0 ) $ !+ T '
+&
&
)
!q ) & 0 *9.1 ) % !+ E (
)
0 ()
!# )( %& 0
!Vk ' $
0.47 0.0005 '
)
!" k ) & 0.0005 *0.002 ) $ !+ Tk '
&
)&
)
+
0
*0.84 ) & !+ Ek )
!qk ) &
) &
(
%
)
0
*0.04 (
!# k ) %
(

Equilibrium Response
Dynamic equation

Equilibrium Response

!!x(t) = F!x(t) + G!u(t) + L!w(t)

At equilibrium, the state is unchanging

0 = F!x(t) + G!u(t) + L!w(t)

Denoting constant values by (.)*

!x* = "F "1 ( G!u * +L!w *)

! t = 0.1sec

Steady-State Condition

Equilibrium Response of
Approximate Phugoid Model

If the system is also stable, an equilibrium point

is a steady-state point, i.e.,
Small disturbances decay to the equilibrium condition

Equilibrium state with constant thrust and wind perturbations

!x P * = "FP"1 ( G P !u P * +L P !w P *)

2nd-order example
System Matrices

Equilibrium
Response

Requirement for
Stability

! f11
F=#
#" f21

! g1 $
! l1 $
f12 $
&; G = #
&; L= #
&
f22 &
g2 &
#
#" l2 &%
%
%
"

# !V *
%
*
%$ !"

" f22 ( f12 %

'
$
%
) l1 ,
" !x1 * %
$ ( f21 f11 '& ") g1 ,
$+
'=(#
$
. !u * + +
. !w *'
+
.
+
.
!x
*
g
l
f
f
(
f
f
(
)
$
'
2
11 22
12 21
* 2 &'
#$
#* 2 &
sI ! F = " ( s ) = s 2 + ( f11 + f22 ) s + ( f11 f22 ! f12 f21 )

#
VN
% 0
&
LV
( = ) %%
)1 VN DV
('
%
gLV
%$ g

&
( + # T* T
( -- %
( , % L* T
( - % VN
(' -. $

&
# DV
(
%
*
( !* T + % )LV
(
% V
$ N
'

/
&
( * -( !VW 0
(
-1
'

= ( s ! #1 ) ( s ! #2 ) = 0

Re ( #i ) < 0

Steady-State Response of
Approximate Phugoid Model

With L!T ~ 0, steady-state velocity depends only on the

horizontal wind
Constant thrust produces steady climb rate
!V * = "
!$ * =

Equilibrium Response of
Approximate Short-Period Model

Equilibrium state with constant elevator and wind perturbations

"1
!x SP * = "FSP
(GSP !uSP * +LSP !w SP *)

L# T
!# T * + !VW*
LV

D (
1%
T# T + L# T V * !# T *
LV )
g '&

Corresponding step response, with L!T = 0

# !q*
%
%$ !" *

&
# L"
M" (
%
(
% VN
% 1 )M ( 1 # M 0 E
&
q
' 33 %
(=)$
2% L
* L"
- 3% ) 0 E
('
M
+
M
VN
q
"/ 3
,+ V
. 4$
N

&
# M"
(
%
*
( !0 E ) % )L"
% V
(
'
$ N

5
&
( * 33
( !" W6
3
(
37
'

Steady-State Response of
Approximate Short-Period Model

Steady pitch rate and angle of attack are not zero

Vertical wind reorients angle of attack

Scalar Frequency Response

with L!E = 0

!q = "

% L#
(
'& V M $ E *)
N

% L#
(
'& V M q + M # *)
N

!# * = "

!$ E *

( M $ E ) !$ E + !# *
W
% L#
(
'& V M q + M # *)
N

Speed Control of
Direct-Current Motor

Characteristics
of the Motor
Simplified Dynamic Model

Angular Rate

Control Law (C = Control Gain)

u(t) = C e(t)
where
e(t) = yc (t) ! y(t)

Rotary inertia, J, is the sum of motor and load

inertias
Internal damping neglected
Output speed, y(t), rad/s, is an integral of the
control input, u(t)
Motor control torque is proportional to u(t)
Desired speed, yc(t), rad/s, is constant

Step Response of
Speed Controller

Model of Dynamics
and Speed Control

Dynamic equation

dy(t) u(t) Ce(t) C

=
=
= [ yc (t) ! y(t)], y ( 0 ) given
dt
J
J
J

1
C
C
u(t)dt = ! e(t)dt = ! [ yc (t) " y(t)] dt
!
J0
J 0
J 0

Step input :
"$ 0, t < 0
yC (t) = #
$% 1, t ! 0

Direct integration of yc(t)

Negative feedback of y(t)

Angle Control of a DC Motor

Control law with angle and angular rate feedback

u(t) = c1 [ y c (t) ! y1 (t)] ! c 2 y 2 (t)

Closed-loop dynamic equation, with y(t) = I2 x(t)

! x!1 (t) $ !
0
&=#
#
'c
!
(t)
x
1 / J
&% #"
#" 2

$ ! x1 (t) $ ! 0 $
&+#
&#
&y
'c2 / J & # x2 (t) & # c1 / J & c
%"
%
"
%
1

! n = c1 J ; " = ( c2 J ) 2! n

where

" = C/J = eigenvalue or

root of the system (rad/s)
# = J/C = time constant of
the response (sec)

Step Response of Angle Controller,

with Angle and Rate Feedback

"$ 0, t < 0
yc ( t ) = #
$% 1, t ! 0

" C%
!$ ' t +
(
!t
y(t) = yc *1 ! e # J & - = yc ()1 ! e.t +, = yc (*1 ! e / +)
,
*)
-,

Integral of the equation, with y(0) = 0

y(t) =

Solution of the integral,

with step command

Single natural frequency, three damping ratios

% Step Response of Damped

Angle Control
F1 = [0 1;-1 0];
G1 = [0;1];
F1a = [0 1;-1 -1.414];
F1b = [0 1;-1 -2.828];
Hx = [1 0;0 1];
Sys1
Sys2
Sys3

= ss(F1,G1,Hx,0);
= ss(F1a,G1,Hx,0);
= ss(F1b,G1,Hx,0);

step(Sys1,Sys2,Sys3)

c1 /J = 1
c2 /J = 0, 1.414, 2.828

Effect of Input Frequency on Output

Amplitude and Phase Angle

Angle Response to a Sinusoidal

Angle Command
Output wave
lags behind the
input wave
Input and output
amplitudes
different
Amplitude Ratio (AR) =

yc (t) = sin ( t / 6.28 ) , deg

y peak
yC peak

"t
Phase Angle = !360 peak , deg
Period

At Higher Input Frequency, Phase

Angle Lag Increases
yc (t) = sin ( t ) , deg

! n = 1 rad / s
" = 0.707

With low input

frequency, input
and output
amplitudes are
about the same
Lag of angle
output oscillation,
compared to input,
is small
Rate oscillation
leads angle
oscillation by ~90
deg

At Even Higher Frequency,

Amplitude Ratio Decreases and
Phase Lag Increases
yc (t) = sin ( 6.28t ) , deg

Angle and Rate

Response of a
DC Motor over
Wide InputFrequency
Range
! Long-term response
of a dynamic system
to sinusoidal inputs
over a range of
frequencies
! Determine
experimentally from
time response or
! Compute the Bode
plot of the system!s
transfer functions
(TBD)

Very low damping

Moderate damping

High damping

Next Time:
Root Locus Analysis