NAVAL POSTGRADUATE SCHOOL

MONTEREY, CALIFORNIA

DT1C
ELECTE
AUG 2 8 1995

THESIS

SIMULINK SIMULATION OF PROPORTIONAL

NAVIGATION AND COMMAND TO LINE OF
SIGHT MISSILE GUIDANCE

by

Patrick Costello

J
U March, 1995

Advisor: Harold A. Titus

Approved for public release; distribution is unlimited.

,,-rffiüTED 9
PTT.fe

19950825 092
 REPORT DOCUMENTATION PAGE Fonn Approved OMB No. 0704-0188

PubUcrq>orting burden for this coDKtion of nrformilion a 1 hour per response, including the time for reviewing instruction, searching existing data
sources, gathering and maintaining Ate data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other
aspect of this collection of information, including suggestions for reducing mis burden, to Washington Headquarters Services, Directorate for Information Operations and
Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188)
Washington DC 20503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE REPORT TYPE AND DATES COVERED
March 1995 Master's Thesis
TITLE AND SUBTITLE SIMULINK SIMULATION OF PROPORTIONAL FUNDING NUMBERS
NAVIGATION AND COMMAND TO LINE OF SIGHT MISSILE GUIDANCE
6. AUTHOR(S) Patrick Costello
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING
Naval Postgraduate School ORGANIZATION
REPORT NUMBER
Monterey CA 93943-5000
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING
AGENCY REPORT NUMBER
11. SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the
official policy or position of the Department of Defense or the U.S. Government.
12a. DISTRffiUTION/AVAIIABIIJTY STATEMENT 12b. DISTRIBUTION CODE
Approved for public release; distribution is unlimited.
13. ABSTRACT (maximum 200 words)
* Proportional Navigation and Command To Line Of Sight missile guidance are explored. A system
flow graph is developed for each guidance technique. The block transfer functions are developed and a
state space representation of the system is defined. The missile systems are then tested using one two-
dimensional engagement and two three-dimensional engagement scenarios. The final three-dimensional
scenario introduces measurement noise in order to evaluate the effect of noise on the guidance
algorithms. The engagement results are then compared to analyze the miss distance of each type of
missile guidance.

14. SUBJECT TERMS Proportional Navigation Missile, Command to Line of Sight 15. NUMBER OF
PAGES *88
Missile, Missile Control.
16. PRICE CODE
17. SECURITY CLASSIFICA- 18. SECURITY CLASSIFI- 19. SECURITY CLASSIFICA- 20. LIMITATION OF
TION OF REPORT CATION OF THIS PAGE TION OF ABSTRACT ABSTRACT
Unclassified Unclassified Unclassified UL
NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std. 239-18 298-102
11
 Approved for public release; distribution is unlimited.

SIMULINK SIMULATION OF PROPORTIONAL NAVIGATION AND COMMAND

TO LINE OF SIGHT MISSILE GUIDANCE

by

Patrick Costello
Lieutenant, United States Navy
B.S.E.E, Marquette University, 1987

Submitted in partial fulfillment

of the requirements for the degree of

MASTER OF SCIENCE ELECTRICAL ENGINEERING

from the

NAVAL POSTGRADUATE SCHOOL

March 1995

Author:

Approved by:

^_ Michael A. Morgan, Chairman

Department of Electrical and Computer Engineering
Accesion For

NTIS CRA&I
DFIC TAB D
Unannounced
Justification

By _
Distribution!

Availability Codes
HI
Avail and/or
Dist Special

f>
 ABSTRACT

Proportional Navigation and Command To Line Of Sight

missile guidance are explored. A system flow graph is
developed for each guidance technique. The block transfer
functions are developed and a state space representation of
the systems is defined. The missile systems are then tested
using one two-dimensional engagement and two three-dimensional
engagement scenarios. The final three-dimensional scenario
introduces measurement noise in order to evaluate the effect
of noise on the guidance algorithms. The engagement results
are then compared to analyze the miss distance of each type of
missile guidance.

IV
 TABLE OF CONTENTS

I. INTRODUCTION 1

II. MISSILE GUIDANCE LAWS 3

A. GENERAL ■ • 3
B. PROPORTIONAL NAVIGATION 4
C. COMMAND GUIDANCE 6

III. SYSTEM DEVELOPMENT 7

A. OVERVIEW 7
B. RADAR DEVELOPMENT 7
1. Proportional Navigation 7
2. Command Guidance 9
C. SEEKER DEVELOPMENT 10
1. Proportional Navigation 10
2. Command Guidance 13
D. GUIDANCE DEVELOPMENT 13
1. Proportional Navigation 13
2. Command Guidance . 13
E. AUTOPILOT DEVELOPMENT I5
1. Proportional Navigation I5
2. Command Guidance I8
F. MISSILE AND TARGET KINEMATICS 19
G. KALMAN FILTER DEVELOPMENT 20

25
IV. SIMULATION RESULTS
25
A. OVERVIEW
25
B. SIMULATION ASSUMPTIONS
26
C. SIMULATION SCENARIOS
1. Constant Velocity In Two Dimensions .... 26
2. Constant Velocity In Three Dimensions ... 26
 3. Three-Dimensional Simulation With Radar Noise
27
D. RESULTS AND SIMULATION COMPARISON 27

V. CONCLUSIONS AND RECOMMENDATIONS 33

A. CONCLUSIONS 33
B. RECOMMENDATIONS 33

APPENDIX 35
A. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 1 . . 35
B. PROPORTIONAL NAVIGATION MISSILE PLOTS FOR SCENARIO 2
40
C. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 2 . . 47
D. PROPORTIONAL NAVIGATION MISSILE PLOTS FOR SCENARIO 3
55
E. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 3 . . 59
F. PROPORTIONAL NAVIGATION SIMULINK MODEL 62
G. COMMAND GUIDANCE SIMULINK MODEL 68
H. MISCELLANEOUS MATLAB CODE 76

BIBLIOGRAPHY 79

INITIAL DISTRIBUTION LIST 81

VI
 I. INTRODUCTION

A guided missile can be controlled using two different

methods. The first is when the missile contains its own
guidance system. This type of missile is beneficial in that
once fired it will track its target. The second type of
guidance has a ground fire control system to command the
missile. This type of missile, called command guided, does
not contain a target seeker. Two radars, or one radar capable
of tracking two targets, are required at the fire control
station; one will track the missile and the other the target.
The fire control system will calculate the required missile
acceleration commands and relay them to the missile by either
a radio link or fiber optic cable.
The type of guidance system implemented is largely
dependent on the missile's mission. A long range missile will
need a self contained guidance system. A point defense
missile will use a self contained seeker or command guidance.
The guidance system supplies the input to the missile
control system. We will use a roll stabilized "skid-to-turn"
missile. The roll stabilization will permit a simpler
analysis because there is no coupling between pitch and yaw.
Figure 1 shows a block diagram for a missile control system.
 Missile Body Seeker

seeker

target
\Jguidance autopilot missile
(

/
tracking
radar

Ground Based Radar

Figure 1. General Missile Guidance System

Missile interception simulations using command to line

of sight and proportional navigation guidance systems are
developed. Chapter II explains the guidance laws. Chapter
III develops the simulation algorithms. Chapter IV tests the
algorithms with known two-dimensional results and a three-
dimensional problem with and without measurement noise.
Chapter V discusses the simulation, conclusions, and
recommendations.
 II. MISSILE GUIDANCE LAWS

A. GENERAL

The missile guidance system provides the autopilot with

the necessary information to produce the required acceleration
commands. The missile/target intercept geometry has several
important parameters. Figure 2 depicts a typical
missile/target intercept scenario.

Figure 2. Missile And Target Intercept Scenario

 Several important parameters can be developed by-
analyzing Figure 2.
Rn, : Tracker to missile range
Rt : Tracker to target range
<7m Tracker to missile line-of-sight angle
°"t Tracker to target line-of-sight angle
Ym Missile velocity vector angle
Yt Target velocity vector angle
The two guidance techniques to be discussed are proportional
navigation and command-to-line-of-sight.

B. PROPORTIONAL NAVIGATION

Proportional navigation missiles are guided by either

reflected or emmitted energy from the target. A passive
missile will be guided from the IR, EO, or RF emmitted by the
target. An active missile will send an RF signal out to track
the target. In each case the energy is received by a seeker
which tracks the target.
Proportional navigation is based on the rate of change of
the missile to target line-of-sight (LOS). The missile
commanded acceleration is proportional to the rate of change
of the LOS. The ratio of the missile turning rate to the LOS
rate of change is called the proportional navigation constant
(N) .

Y
ö
v <21>
N- —- N'
Vm

The proportional navigation constant must be greater than 2 to

ensure system stability. A proper value of N will ensure that
the missile to target angle amt will remain constant thus
ensuring missile intercept. Figure 3 illustates this point

.Impact

TARGET FLIGHT

MISSILE FLIGHT
Figure 3. Missile Collision Course Theory

Assuming an acceleration is applied at right angles to

the velocity vector of the missile for a period of time dt,
the missile's velocity will then be Vm(t+dt). The velocity
vector will have changed direction by dYm-
Assuming a small angle approximation yields

a dt = V dv
(2.2)
a = vy

Combining (2.1) and (2.2) leads to

a m = VNÖ
m
(2.3)

This result is the proportional navigation law that will be

implemented in this simulation.

C COMMAND GUIDANCE

The Command To Line Of Sight (CLOS) missile is given

guidance commands that keep the missile in the LOS between
the launch point and the target. The distance between the
missile and the LOS is defined as the cross range error
(CRE). The fire control system will supply the proper
commanded acceleration to drive the CRE to zero.
Since two separate radars are required for this type of
guidance the problem geometry is slightly different than
previously described. Figure 4 shows the CLOS system
geometry.

2;

MISSILE

TARGET
CREfct*
\CRE
\ \
\ \ ^
\ \ ^^^
\\ ^^

^^ 1
1^7* ] 1
''jOtapitch/ |
y
4—-^/ <Ttpit<Jh 1
yaw
^^ \cRE(i»
/ Ota yaw

Figure 4. Command To Line Of Sight Geometry

 III. SYSTEM DEVELOPMENT

A. OVERVIEW

The system block diagram is shown in Figure 1. The block

transfer functions, system dynamics, and simulation equations
will be developed for the simulation.

B. RADAR DEVELOPMENT

Target flight is tracked using angles in the pitch and

yaw planes. The pitch plane is defined as the vertical plane
that contains the target and the radar. The yaw plane is
defined as the xy plane.
1. Proportional Navigation
Proportional navigation system geometry is shown in
Figure 5.

2r

MISSILE^^
/[ ' —— __TARGET

^^T^a<f «Ttpitcjh |
-^Km " —■ ■—i 1
^^ "~^ 1
^-—^ ^ 1

/ Cmyaw 7* ^^ 1
*-^
Gtyav "^ ^ -

X ^ 1

Figure 5. Proportional Navigation System Geometry

From Figure 5 the following angles can be defined:
■'myaw Missile yaw angle
J
mpitch Missile pitch angle
y
tyaw Target yaw angle
a
tPitch : Target pitch angle
The system requires that the following ranges be defined:
Rn, : Radar to missile range
Rt : Radar to target range
R : Missile to target range
By applying elementary trigonometry to the cartesian
system geometry defined in Figure 5, the following equations
can be derived

= arctan
"V«

Z
m
a = arctan
'"pitch 2
Jx
\ V m+y2 m/

arctan
\xtJ
(3.1)

c
arctan
pi tab
/*\*y\,

Rm . VJx2 ♦ y2 + z2

2
R
■fc • y.2 • *,'

*-VK-*.>2 * (ye-y->2* <v2.>

The radar system will produce the following angles
a„yaw
=„, : Missile to target yaw plane angle
■'pitchMissile to target pitch plane angle
The angles are given by the equations

ayaw = arctan

(3.2)

°Pitch = arctan
W(xt-*J2+<yt-yJ2J
The radar will send these angles to the respective yaw and
pitch seeker elements.
2. Command Guidance
The CLOS radar will produce a cross range error signal
and relay this signal to the missile autopilot. The cross
range error is the distance between the missile and the radar
to target LOS. Figure 4 shows the CLOS geometry.
From Figure 4 and vector calculus the cross range error
of the missile can be defined as follows

\R x R\
CRE\ = 1-SL Si (3.3)

This calculation yields the following equation

\CRE\ = J(y z-y.z )2+(xz -x zj2+(x v-xjy ) (3.4)

The missile autopilot requires that the cross range error

be broken into the yaw and pitch components. Analyzing Figure
4 yields the following equations
 CREyaw = Jx 2+y 2.sin(o\ - c )
2 (3.5)
CREpitch = JCRE
v
-CREya"2 sign (a t - a"Vlteh )
pitch

The sign function ensures that the pitch plane cross range
error can be positive or negative. A positive cross range
error indicates that the missile is leading the LOS. A
negative cross range error indicates the missile is trailing
the LOS.

C. SEEKER DEVELOPMENT

1. Proportional Navigation
The seeker for proportional navigation measures the rate
of change of the missile to target LOS angle. A simple
gimballed seeker will use the angular rate of change of the
seeker head as an estimate of the rate of change of the LOS
angle. Figure 6 shows the seeker.

Seeker

Target

Figure 6. Seeker Head Model

10
 The equation of motion of the seeker head will yield the
estimate of the angular rate of change of the LOS. The seeker
head equations for the pitch and yaw planes will be identical.
We will develop the equations for only the yaw plane. The
equation of motion for the seeker head is given by

where
T = Torque applied to the seeker head
I = Moment of inertia of the seeker head
ß = Seeker bore sight angle
Solving (3.6) yields

ß = - = -^(ß-c) - k2$ = -kß-kfi+k^ (3.7)

Taking the Laplace transform of (3.7) gives

s2(3(s) = -k2s$( s)-kfi(s)+k1a (s) (3.8)

Then we solve for the seeker transfer function

ß(s) *i _ *i
2
O(s) ~ s +k02s+k,1 ' [ e
S+
1 )2 (3-9>
{ M
where TSH is the seeker head time constant.
A typical seeker head time constant is TSH= 1/8, using
this value produces the following constants

11
 2
( \
]r -
( 1 ) 1
= 64
(3.10)
1 ) 2
2 16
\\ T SB) .125

The signal flow graph, using these constants, can be seen

in Figure 7.

Figure 7. Proportional Navigation Seeker SFG

The following state space representation can then be

implemented

0 1 0
x SB xm
SH
+ u SH
-64 -16 64
X. (3.11)
X
SH
x„

u SH = a

12
 2. Command Guidance
The CLOS missile control system does not contain a seeker
head. All missile control functions are processed and
developed by the fire control system located at the radar
site.

D. GUIDANCE DEVELOPMENT

1. Proportional Navigation
The missile guidance system implements the proportional
navigation law explained in Chapter II. The major difference
is that an estimate of the angular LOS rate is used vice a
measurement of the actual LOS rate. Therefore, the rate of
change of the missile's velocity vector is given by

y =N$ (3.12)

This leads to the following state variable representation

1 N 0 ft
r
in pi tch pi tch
= (3.13)
* m yaw
0 N r
yaw

2. Command Guidance
The guidance for a CLOS missile is developed from the
rate of change of the missile's cross range error. The
missile acceleration is equal to the rate of change of the
cross range error. This rate of change is then used as a
commanded acceleration in the autopilot.
The commanded acceleration is developed to provide good
missile response. To ensure good response the missile
acceleration must be of the form

s2 ♦ «x + ß)s ♦ aß <3-14>

13
This will provide the damping necessary for the missile to
perform correctly.
Using equation (3.14) the following commanded
acceleration is developed

a = CRE • OtCRE * pCRE (3.15)

Taking the Laplace transform of (3.15) yields

and
= s* + as + p (3.16)
CRE(s)

a = 40 and p = 196 produces two real roots at s=-5.7l7l and

s=-34 .2829.
The signal flow graph for the guidance system is shown in
Figure 8.

Figure 8. Command To Line Of Sight Guidance SFG

A state space representation of the guidance system is

14
 CREpitch
.„ „

and."pitch 196 40 0 0 CREpitch

.„ h
(3.17)
and.. 0 0 196 40 CREyaw

CREyaw
E. AUTOPILOT DEVELOPMENT

1. Proportional Navigation
A simple autopilot can be developed by applying a torque
about the center of gravity of the missile. Analyzing the
equation of motion
I
crX
CG ' m (3.18)

and noting that this must also satisfy equation (3.14) to

achieve stable performance, yields the following relationship

Y = = -KV + KN$" (3.19)

CG

Taking the Laplace transform of (3.19) yields

VJs) KN
(3.20)
$(s) S + K

and defining TAP as the autopilot time constant produces

(3.21)
AP

The signal flow graph for the autopilot, with k=l, is shown in
Figure 9.

15
 Figure 9. Proportional Navigation Autopilot SFG

The state space representation can be written as follows

"bitch -1 0 "bitch 1 0
uAP
0 -1 0 1
m
yav "W
(3.22)
N 0 & "bitch
u AP
0 N
mym

The missile acceleration commands can be derived by-

looking at the missile's velocity vectors. Figure 10 shows
the two-dimensional missile acceleration geometry.

16
 Figure 10. Missile Acceleration Geometry

It can be shown that the velocity in the pitch and yaw

planes is given by

= v cos (y - a )
(3.23)
V'"pitch = V m COS(Y
'"pitch
)

The acceleration components are then a function of the angular

rate of change of the velocity vector

a - v y
"pitch "pitch "pitch
(3.24)
a - V V

The angular acceleration commands are then distributed to the

missile's cartesian coordinate accelerations using the
following geometric relationships

17
 x'"■pitch =-a "Utah sin(apitch'
._ J cos (ayaw')
y"tltch =-a "feltah sin (apitch'
.^ J sin(Gyawr')
Z = a COS(G .„.) (3 25»
"titeh «feiteb pitch' U.^Of

xm = -a sin(aya,,r )
y.» v
y = a cos(o )

The missile acceleration in each plane is then

X = X + X
m m
'"pitch yv

= +
y'm m y'm.
"pitch
^myw
m (3.26)
'

z = z

and the total missile acceleration is

«. ■ K * n ♦ *i (3 27)
-

2. Command Guidance
The CLOS autopilot also takes the guidance commands and
translates them into missile accelerations. Similar to the
proportional navigation autopilot, this autopilot translates
the angular accelerations into cartesian coordinate
accelerations.
The commanded angular accelerations of equation (3.15)
are translated to cartesian accelerations using the following
relationships

18
 = _a
x'm"Vitch —w
«^iteh
sin (apatch'
.„ Jcos (ayaw')

y!„Vtch ■ _a
^w
^itch
sin(GPitch'
.„ Jsin(Gyaw')

z"Wtch = a «""«Wee*
, cos (aPitch'
.„ .) /3 28^
U.^ÖJ

xi^na, = -a cmdy.,,
_, sin(Gyaw )

y = a , cos (a )

The overall missile accelerations are also given by equations

(3 .26) and (3.27) .

F. MISSILE AND TARGET KINEMATICS

The missile and target kinematics can be developed using
the state space representation

x=\xxyyzz\
(3.29)
x z
t - K *t yt yt t *J
The system is then represented by

xm = Am xwi + Bin u in
(3.30)
x A x + B u
t = t t t t

19
where

0 10 0 0 0
0 0 0 0 0 0
0 0 0 10 0
0 0 0 0 0 0
0 0 0 0 0 1 (3.31)
0 0 0 0 0 0
u = fx ym z m
u
t = K y't \

A signal flow graph for the missile and target kinematics can
be seen in Figure 11.

1/s 1/s
Baäd^ Xm
—►— -9— —► •
(a) Missile Kinematics SFG

1/8 1/s
a m- k w xt
m
r f

(b) Target Kinematics SFG

Figure 11. Missile And Target Kinematics

G. KALMAN FILTER DEVELOPMENT

The introduction of noise into the simulation creates a
more realistic scenario. The problem is to determine the
target's flight path by filtering the noise. This simulation
uses an extended Kaiman filter to estimate the target's

20
flight.
The noisy observed target flight is the input to the
filter. The cartesian and spherical coordinates of the target
are then used in the Kaiman iteration to estimate the target's
position. The filter is developed to use preprocessed linear
pseudomeasurements. These measurements are given by

x(kT)
\ (tan2atan2ß + tan2a ♦ tan2ß+l)
2 2
y(kT) = r tan a (3.32)
\ (tan atan ß + tan2a ♦ tan2ß+l)
2 2

r2tan2ß
z(kT)
\ (1 ♦ tan2ß)

where
a = LOS pitch angle
ß = LOS yaw angle
The measurement equation then becomes

10 0 0 0 0
y(kT) 0 0 10 0 0 x(kT) ♦ v (3.33)
0 0 0 0 10

where Vk = N(0,R), and R = H (kT) R*HT(kT) . H(kT) and R* are

given by

21
 5x(r,a,ß) 5x(r,a,ß) 5x(r,g,ß)
5r 5a öß
öy(r,a,ß) öy(r,a,ß) Sy(r,a,ß)
H(kT) =
ör 6a öß
öz(r,a,ß) öz(r,a,ß) öz(r,a,ß)
(3.34)
5r 5a öß
a2
r 0 0
R' = 0 Ga2 0

0 0 GB2

The discrete time system model then becomes

x( (x+l)T) - Fx(kT) + W„

Wk * N(0,Q)

z 0 0
0 2 0
0 0 s (3.35)

gT3 gT'
3 2
gT 2
gT

The initial condition fox the filter is

x(0) « N(x0,P0) (3.36)

22
The Kaiman algorithm is then given by the following set of
equations
xk~l/k = Fx
X eX
k/k

p FP FT
k.m= kn +Q
K
k= *»V*
BT BP
I M
HT
* RJ <3-37>
Jt.l/JM. Jt»l/Jt kLJ ^ k*l/kJ

T
L J l J
Jt»l/Jfe»l k Jt.l/Jt k Jt Jt Jt

23
24
 IV. SIMULATION RESULTS

A. OVERVIEW

The proportional navigation and CLOS simulations are

tested using three target flight scenarios. In the first
scenario the target has constant velocity and level flight in
two dimensions. In the second, the target has a constant
velocity and level flight in three dimensions. Finally, in
the third, noise is added to the three-dimensional scenario.
The Simulink models and associated MATLAB code for the
proportional navigation and CLOS simulations are contained in
the Appendix.

B. SIMULATION ASSUMPTIONS

The following assumptions are held throughout the

simulation:
(1) Acceleration due to gravity does not effect the
missile or the target.
(2) The missile is lying in the xy plane at launch.
(3) Missile acceleration is limited to 30g.
(4) The proportional navigation constant is N=6.

25
C. SIMULATION SCENARIOS

1. Constant Velocity In Two Dimensions

The first scenario is a two-dimensional engagement. The
target is flying at a constant altitude with no acceleration.
The target parameters are as follows
x
t •. 30000 ft
■ -3000 ft/s
*t ■

*t ■
. 0 ft/s2
yt-■ 0 ft
yt ■• 0 ft/s
y't ■. 0 ft/s2
z ■ 1000 ft
t •
±
t ■
• 0 ft/s
2
*\ •. 0 ft/s

2. Constant Velocity In Three Dimensions

The next scenario is a three-dimensional engagement. The
target is flying at a constant altitude with no acceleration.
The target parameters are as follows
X ■ ■ 60000 ft
X ■ ■ -2121 ft/s
X ■■
0 ft/s2
yt'■ 10000 ft
yt ■• -2121 ft/s

y't ■
■ 0 ft/s2
z
t'
. 1000 ft
z
t'
■ 0 ft/s
2
z't ■. 0 ft/s

26
 3. Three-Dimensional Simulation With Radar Noise
The final simulation uses the same target parameters as
the three-dimensional constant velocity simulation. White
noise is added to the target flight. This simulates received
noise in the target's radar return. The noise has the
following characteristics
ar - 15 ft
ayaw - 1°

D. RESULTS AND SIMULATION COMPARISON

Figure 12 indicates the missile leads the target. This
is attributed to the slow missile autopilot time constant
(1^=1 sec) and the target's speed advantage of mach 3 to mach
2 over the missile. This problem is exaggerated in figures 12
and 13 since the z scale is twenty times the x scale. It was
determined by considering the z acceleration profile in figure
16, the z velocity profile in figure 19, and the z position
profile in figure 12, that this effect was caused by the
autopilot.
Figure 14 shows the rate of change of o is positive for
approximately 1 second; thereafter it is negative but, for 2
seconds the missile has a positive commanded acceleration.
Figures 15 and 16 show the missile's acceleration variations.
Figures 17, 18 and 19 show the missile's velocity variations.

27
 Missileflarget Engagement in the xz Plane

0.5 1 1.5 2 2.5

Distance (ft x1e4)
x10
Figure 12. Proportional Navigation Scenario 1. Missile and
Target Trajectories in the xz Plane.

Missile/Target Engagement

~ 1000

104
Distance (ft)) -1 0 Distance (ft x 1e4))

Figure 13. Proportional Navigation Scenario 1. Three-

Dimensional Engagement Plot.

28
 LOS Angle Sigma in Ine Pitch Plane
0.1

0.08

0.06

2 3 4
Time (seconds)
Figure 14. Proportional Navigation Scenario 1. LOS Angle
Opitch.

Commanded Acceleration in the x Direction

St
'S"
a> -5c
CO

-10

-15

-20

-25
2 3 4
Time (seconds)
Figure 15. Proportional Navigation Scenario 1. Commanded
Acceleration in the x Direction.

29
 Commanded Acceleration in the z Direction
300

200

100

El °
g"-100

-200 N.

-300 ^—J ■

-400
() 1 2 3 4 5 (
Time (seconds)
Figure 16. Proportional Navigation Scenario 1. Commanded
Acceleration in the z Direction.

Missile Velocity
2170

2165 S—N. /'

2160 / \ '
2155 f \ / ■
2150

S2145

2140

2135 / ^~~*^
2130 I

2125

2120
(} 1 2 3 4 5 (
Time (seconds)
Figure 17. Proportional Navigation Scenario 1. Total Missile
Velocity.

30
 Missile Velocity in the x Direction
2150

2145-

2140-

2 3 4
Time (seconds)
Figure 18. Proportional Navigation Scenario 1. Missile
Velocity in the x Direction.

Missile Velocity in the z Direction

400

2 3 4
Time (seconds)
Figure 19. Proportional Navigation Scenario 1. Missile
Velocity in the z Direction.

31
 Plots for the other scenarios are given in the Appendix.
The following table summarizes the missile's closest point of
approach (CPA), and the time of the CPA for each simulation.

Scenario Simulation CPA Time of CPA

1 Prop Nav 4.13 ft 5.89 s
CLOS 1.39 ft 7 .18 s
2 Prop Nav 14.94 ft 14.72 s
CLOS 1.24 ft 19 .51 s
3 Prop Nav 27.15 ft 14 .5 s
CLOS 267.79 ft 22.34 s
Table 1. Missile Miss Distance Summary

Overall, the proportional navigation missile achieves a

quicker target intercept time. The miss distances for each
missile are very close, except when noise is added. The CLOS
missile degrades significantly in the presence of noise.
The proportional navigation missile is a superior
missile. The CLOS missile is unable to give satisfactory
results when sensor noise is added to the simulation. For
very short range intercept scenarios, where sensor noise is
negligible, the missile will perform well. The proportional
navigation missile will perform well for any engagement
scenario. This fact makes proportional navigation preferable
for missile guidance.

32
 V. CONCLUSIONS AND RECOMMENDATIONS
A. CONCLUSIONS

The simulation provides insight in chosing the proper

type of missile guidance. The two types of guidance explored
both give acceptable miss distances without sensor noise
However, when sensor noise is present the proportional
navigation missile outperformed the CLOS missile.
The presence of an on board seeker gives the proportional
navigation missile an advantage when dealing with sensor
noise. since the sensor is on the missile as it closes the
target, the sensor noise will have less of an effect on the
detection of the target. The CLOS missile is guided from a
stationary radar at the launch site. The error incurred from
sensor noise does not decrease as the missile approaches the
target. To overcome this problem the CLOS missile will
require a very sophisticated tracking radar that has very
little sensor noise.
The addition of noise to the engagement provides a more
realistic scenario for the missile control problem.
Developing a noise filter and adjusting the missile
characteristics to adapt to the noise created a unique and
educational challenge. The increased realism reinforced the
fact that actual missile control developement is a compromise
of design requirements.

B. RECOMMENDATIONS

The simulation can be taken to several different levels.

The target flight can be modified for different engagement
scenarios. A manuevering target would provide another level
of realism to the engagement.
An adjoint model could be built for each simulation.

33
This would aid in the miss distance analysis for the two
missiles.
Finally, different noise filters can be developed and
tested. The miss distance will be decreased if better noise
filtering is achieved during the simulation.

34
 APPENDIX

A. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 1

Missile/Target Engagement in the xz Plane

1200
Target
1000

800

600
3w

400

200

1 1.5 2 2.5
Distance (ft x1e4) x10

Figure 20. Command Guidance Scenario 1. Missile and Target

Trajectories in the xz Plane.

35
 Missile Target Engagement

Distance (feet) -1 0 Distance (feet x 1 e4)

Figure 21. Command Guidance Scenario 1. Three- Dimensional

Plot.

Cross Range Error

3

2.5

2 -

« 1.5 -

0.5

0
() 1 2 3 4 5 6 7 1i
time (sec)
Figure 22. Command Guidance Scenario 1. Cross Range Error.

36
 Commanded Acceleration in the x Direction
10

Or
-10

-20

-50
-60
-70
-80
-90
1 3 4 5 8
time (sec)
Figure 23. Command Guidance Scenario 1. Commanded
Acceleration in the x Direction.

Commanded Acceleration in the z Direction

1000

800

600

IF 400
CO

200

-200
3 4 5
time (sec)
Figure 24. Command Guidance Scenario 1. Commanded
Acceleration in the z Direction.

37
 Missile Velocity
2125

2124

2123

2122

I
SB
2121
2120

2119

2118

2117
1 3 4 5 8
time (sec)
Figure 25. Command Guidance Scenario 1. Total Missile
Velocity.

Missile Velocity in the x Direction

2122

2120

2118 -

2116
^^^\
o2114
e
2112 : ^"^\

2110 \.

2108 \
2106 \

2104
() 1 2 3 4 5 6 7 I$
time (sec)
Figure 26. Command Guidance Scenario 1. Missile Velocity in
the x Direction.

38
 Missile Velocity in the z Direction
300 ill 1 1 1 I

250

200 /

|l50

100

50

0
(3 12 3 4 5 6 7 8
time (sec)

Figure 27. Command Guidance Scenario 1. Missile Velocity in

the z Direction.

39
B. PROPORTIONAL NAVIGATION MISSILE PLOTS FOR SCENARIO 2

Proportional Navigation MissileAarget Engagement Scenerio 2.

Distance (ft x1e4) -2 0 Distance (ft x1e4)

Figure 28. Proportional Navigation Scenario 2. Missile and

Target Trajectories in Three Dimensions.

40
 x -j 04 Missile/Target Engagement in the xy Plane

0.5

ä 0
?"
*-0.5 \ Missile /
0)
1 \ Aarget
I -
Q
-1.5

-2

-2.5
(3 12 3 4 5 6
Distance (ft x1e4) x104

Figure 29. Proportional Navigation Scenario 2. Missile

Target Trajectories in the xy Plane.

Missile/Target Engagement in the xz Plane

1200

Target
1000-

2 3 4
Distance (ft x1e4)
x10
Figure 30. Proportional Navigation Scenario 2. Missile and
Target Trajectories in the xz Plane.

41
 Missile/Target Engangement in the yz Plane
1200

1000

800
Qi
^
X
*s
<u 600
u
c
2
CO
D 400

200-

-2.5 -1.5 -1 -0.5 0 0.5 1

Distance (ft x1e4)
x10
Figure 31. Proportional Navigation Scenario 2. Missile
Target Trajectories in the yz Plane.

LOS Angle Sigma in the Pitch Plane

025

02

~ 0.15
ID

2, 0.1 ■

CU
CO

* 0.05

0
" '
J
-0.05
(D 5 10 15
Time (sec)
Figure 32. Proportional Navigation Scenario 2. LOS Angle
Opitch.

42
 LOS Angle Sigma in the Yaw Plane
02

0 >

-02 ■ -

?
.2-0.4 - •
2
£-0-6
'
-0.8 - -

-1 - •

5 10 15
Time (sec)
Figure 33. Proportional Navigation Scenario 2. LOS Angle
Oyaw.

Commanded Acceleration x Direction

500

400

300

If
«o.
200

100

-100 15
5 10
Time (seconds)
Figure 34. Proportional Navigation Scenario 2. Commanded
Acceleration in the x Direction.

43
 Commanded Acceleration y Direction
100

8"-200

-300

-400

-500
5 10 15
Time (seconds)
Figure 35. Proportional Navigation Scenario 2. Commanded
Acceleration in the y Direction.

Commanded Acceleration z Direction

160

140

120

100
i\
i \ -
8 \
P °
IT 60 -
in

* 40 \ i
20 \ j
0
\ ^-^
-20 x^_ ___^-^^^^
-40
C) 5 10 15
Time (seconds)
Figure 36. Proportional Navigation Scenario 2. Commanded
Acceleration in the z Direction.

44
 Missile Velocity
3000 -

2900 I ■

2800 ■

2700- ■

o 2600

«2500 ■

2400 ■

2300 ■

2200 ■

2100 15
) 5 10
Time (seconds)
Figure 37. Proportional Navigation Scenario 2. Total Missile
Velocity.

Missile Velocity in the x Direction

2140 ■ 1 '

2120 -\

2100 " \

2080 \

5 2060 \

2040 \

2020 \ ■

2000 ^\^^ )

198C 15
0 5 10
Time (seconds)
Figure 38. Proportional Navigation Scenario 2. Missile
Velocity in the x Direction.

45
 Missile Velocity in the y Direction

-500-

-1000-
o

-1500-

-2000

-2500
5 10 15
Time (seconds)
Figure 39. Proportional Navigation Scenario 2. Missile
Velocity in the y Direction.

Missile Velocity in the z Direction

200

150

100
u

-50
5 10 15
Time (seconds)
Figure 40. Proportional Navigation Scenario 2. Missile
Velocity in the z Direction.

46
C. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 2

Command To Line Of Sight Missile/Target Engagement Scenerio 2

1000

800

. 600
o
| 400

200

x10

Distance (ft x1e4) -4 0 Distance (ft x1e4)

Figure 41. Command Guidance Scenario 2. Missile Target

Engagement.

47
 x 1 o4 Missile/Target Engagement in the xy Plane

2 3 4
Distance (fix 1e4)
x10
Figure 42. Command Guidance Scenario 2. Missile Target
Trajectories in the xy Plane.

Missile/Target Engagement in the xz Plane

2 3 4
Distance (ft x1e4)
x10
Figure 43. Command Guidance Scenario 2. Missile Target
Trajectories in the xz Plane.

48
 Missile/Target Engagement in the yz Plane
1200

Target
1000

800

0)
o
c 600 Missile
8CO

400

200

-2 -1
x10 Distance (ft x1e4)
Figure 44. Command Guidance Scenario 2. Missile and Target
Trajectories in the yz Plane.

Cross Range Error

50 -
'
45 -

40 -

35 ■
Distance (ft)
o
Ol
O

15

10 A

5
\ ___——'^-— '
0 I 5 10 15 20
C
Time (seconds)
Figure 45. Command Guidance Scenario 2. Missile Cross Range
Error,

49
 Cross Range Error in tfie Pitch Plane

10
Time (seconds)
Figure 46. Command Guidance Scenario 2. Cross Range Error in
the Pitch Plane.

Cross Range Error in Ihe Yaw Plane

ou
t\t\ - .

40 -

30 .

f20
I
a»
Ö
10

0 1 __^^_
-10L
0 5 10 15 20
Time (seconds)
Figure 47. Command Guidance Scenario 2. Cross Range Error m
the Yaw Plane.

50
 Commanded Acceleration in the x Direction
200

L
-200

El
1" -400
jo.

-600

-800

-1000
II 20
10 15
Time (seconds)

Figure 48. Command Guidance Scenario 2. Commanded

Acceleration in the x Direction.

Commanded Acceleration in the y Direction

IUUU :

800- -

600 ■

400 -

20
p °-
? 0-
42.
fe
-200
/
-400

-600

-800 .

-1000 > 5 10 15 2»0

c
Time (seconds)

Figure 49.
Acceleration in the y Direction.

51
 Commanded Acceleration in the z Direction
mnn

800 - ■

600 .

El
¥ 400
t
200

0 v~- \^

-200
t) 5 10 15 20
Time (seconds)
Figure 50. Command Guidance Scenario 2. Commanded
Acceleration in the z Direction.

Missile Velocity
3400
/
3200 A /

3000 / ^^^^
2800 /
«2600
e
2400 /

2200 /

2000 i /
1800

1600
(} 5 10 15 20
Time (seconds)
Figure 51. Command Guidance Scenario 2. Total Missile
Velocity.

52
 Missile Velocity in the x Direction
trnn

2000
V
1500

1000 •

o 500

** 0

-500

-1000 -

-1500

-2000
(D 5 10 15 20
Time (seconds)
Figure 52. Command Guidance Scenario 2. Missile Velocity in
the x Direction.

Missile Velocity in the y Direction

500

0 ^s.

-500 N.

-1000 \
u
a>
J»
£
-1500
\

-2000

-2500 \

-3000
I3 5 10 15 20
Time (seconds)
Figure 53. Command Guidance Scenario 2. Missile Velocity in
the y Direction.

53
 Missile Velocity in the z Direction

10 15 20
Time (seconds)

Figure 54. Command To Line Sight Scenario 2. Missile

Velocity in the z Direction.

54
D. PROPORTIONAL NAVIGATION MISSILE PLOTS FOR SCENARIO 3

Missile Target Engagement

x10
Distance (ft x1e4) -4 0
Distance (ft x1e4)

Figure 55. Proportional Navigation Scenario 3. Missile and

Actual Target Trajectory.

55
 Missile Target Engagement

2000

^1500

8c
1000
B
w
Q 500

x10

Distance (ft x1e4) Distance (ftx1e4)

Figure 56. Proportional Navigation Scenario 3. Missile and

Filtered Target Trajectory.

Missile and Actual Target Engagement in the xz Plane

1 2 3 4 5 6
Missile and Filtered Target Engagement in the xz Plane x io"
2000

~\v4l
3 4
Distance (It x1e4) JSJSL
Figure 57. Proportional Navigation Scenario 3. Missile and
Target Trajectories in the xz Plane.

56
 x 1 o4 Missile and Actual Target Engagement in the xy Plane

12 3 4 5 6
4 4
x ! o Missile and Filtered Target Engagement in the xy Plane x 1 Q

3 4
Distance (ft x1e4)
JOE
Figure 58. Proportional Navigation Scenario 3. Missile and
Target Trajectories in the xy Plane.

Missile and Actual Target Engagement in the yz Plane

4cnn
" 1 1 I 7 1
c
c
C
ance (ft)
o
o
O

^\
«2 500
O >v
0 1 1 1 1 1 1

-2 5 -2 -1.5 -1 -0.5 0 0.5 I

Missile and Filtered Target Engagement in the yz Plane x 1 rj4
tnnr\

gl500 r
a>
c 1000
CD ^"""''"■•^^^"'^"""^'■vs^r^
CO
Q 500 >^
■ i i i i i i
0
-2 5 -2 -1.5 -1 -0.5 0 0.5 1 1 5
D istance (ft x 1 e4) x 10^

Figure 59. Proportional Navigation Scenario 3. Missile and

Target Trajectories in the yz Plane.

57
 Missile Velocity
3000

2900

2800 /
2700 /
u2600
CD
/
^2500

2400 /

2300 /
2200
91 nn " ■ ' ■ i i

5 10 15
Time (sec)
Figure 60. Proportional Navigation Scenario 3. Total Missile
Velocity.

Missile Acceleration in the xDirection

500 , ,
CM
<
0)
CO
0

-500 1 I

0
i 5 10 15
Missile Acceleration in the y Direction
CM
500 i i
<
a> 0
to
">s^-^. ^—-—-~~"~""^ >
■ ■
-500
0l 5 10 15
Missile Acceleration in the z Direction
CM
200 ■ i
<
a>
«o
0
■
-200 >

10 15
Time (sec)
Figure 61. Proportional Navigation Scenario 3. Missile
Accelerations.

58
E. COMMAND GUIDED MISSILE PLOTS FOR SCENARIO 3

X10
Distance (ft x1e4) -4 o Distance (ft x1e4)

Figure 62. Command Guidance Scenario 3. Missile and Target

Trajectory.

59
 4
x! o Missile and Actual Target Trajectories in the xy Plane
^ 2

1°
I"2
CD

0 12 3 4 5 6
4
x 10 Missile and Filtered Target Trajectories in the xy Plane x io"
^ 2

* 0
a>
1-2
as
to

3 4 5
Distance (ft x1e4) X1Q"
Figure 63. Command Guidance Scenario 3. Missile and Target
Trajectories in the xy Plane.

Missile and Actual Target Trajectories in the xz Plane

1500

^1000 -■■

o
c
CO
« 500
O

1 2 3 4 5 6
4
Missile and Filtered Target Trajectories in the xz Plane x -JQ
2000
gl500

3 4
Distance (ftx1e4)
*1Q
Figure 64. Command Guidance Scenario 3. Missile and Target
Trajectories in the xz Plane.

60
 Missile and Actual Target Trajectories in the yz Plane
1500

1000
ü
C
«5 500
5
-3 -2-1 0 1
4
Missile and Filtered Target Trajectories in the yz Plane x IQ
2000
:i5oo
1000
M
Q 500

-2 -1 0
Distance (ft x1e4) _xJÜ
Figure 65. Command Guidance Scenario 3. Missile and Target
Trajectories in the yz Plane.
Cross Range Error
g400

CD
£200
**
W
b o*- 25
5 10 15 20
Cross Range Error in the Yaw Plane
g 500

b-500
5 10 15 20 25
Cross Range Error in the Pitch Plane
g 501 1 1 —i 1

b-50
10 15 20 25
Time (sec)
Figure 66. Command Guidance Scenario 3. Cross Range Error.

61
F. PROPORTIONAL NAVIGATION SIMULINK MODEL

©-
Clock To
]
Workspace5

(ET <=°»iu[ii-u[2]|-r .
f (u)
sig_p
5-*&>*!K]
Control
53:
Seeker
Missile

V3J 3-
Pitch Autopilot
Pitch

■3
Fight
x

CPA Calculation
Target
Flight

Muxl
55-1
»I flu) \ ^U^L>^^ cos(u)
Missile
Flight
Y
Control
Seeker
Yaw Autopilot
Yaw

5- Max Step .001

Min Step .001
:>■

CPA Calculation Tol le-3

and Missile
Stop Simulation Flight
Z

Figure 67. Proportional Navigation Missile Model.

62
 MMLAB
I-H3
x actual
phi
Function
Kaiman
Kaiman
Estimate
Filter
x target pos
actual
>121 1/s
Vxt xt(0) x target pos
HZ]
y actual
observed

EHE—
Vyt
HL3-
yt(0) yta
-♦
3> ^ yt
y target pos
y target pos observed
Cartesian Spherical
actual to to
Spherical Cartesian
z target pos

is
observed
z actual

z target
actual

Figure 68. Target Flight with Noise,

63
 -2121 l/s 'W 1
Vxt xt(0 ) xt
L-H xt |
x pos target
-2121 1/a 2
Vyt yt(0 ) yt
1 "vt 1
i y^ i
1000 y pos target
3
z pos zt
t t -r
n .... 1
z pos t arget

Figure 69. Target Flight without noise.

betadot

m-
ngnu =tE>-H£>—;fl3 >©-
Sum2 64 Sum be^dot beta

16

Figure 70. Proportional Navigation Pitch and Yaw Seeker,

64
 r-*0
gammadot
-m
Q-
N betadot S3—•> 1/s
gammadot
■0

■o-

Figure 71. Proportional Navigation Pitch and Yaw Autopilot

Aux
0~~X
si
9_P
Anx »| f(u)
—sin(u[l])cos(u[2 )

»I acmdx |
2 tei command
amp accel x

vmx

\=t^ 1/s 1/s ♦UJ

X
Sum -966 vmx(O) -350C xm(0)
966 JOUU

amy
HE?
I vx \*1 I xm P
velocity x pos missile
missile
x

-Hsi n[u)H
sig_y sin(u)

Figure 72. Proportional Navigation Missile Flight in the x

Direction.

65
 sig_y cos(u)

=ttZh
amp ■*|acmdyj
command
accel y
vym

l/s|—f^H
Sum ym
amy r*V-" -966 vym(0
vym(O) -3500 ym(0)
* 966 3500

I yy H ' I ym
.sin(u[2])sin(u[l]) y velocity y pos missile
sig_p missile

Figure 73. Proportional Navigation Missile Flight in the y

Direction.

* acmdz
amp command
accel z r*H
►0
-966 vz(0) -3500 zm(0)
966 3500
2 f (u)
sig_p cos(u)
z
z velocity Pos missile
missile

Figure 74. Proportional Navigation Missile Flight in the z

Direction.

66
 H f(u) ► cpa
Missile sqrt( (u[l]-u[4])A2 CPA of Target
and + (u[2]-u[5])*2 with Noise
Target +u[3]-u[6])*2)
Flight

f (u)
A
sgrt(u[l]
A
2
(■u[2] 2+u[3]*2)

Constant

STOP

f (u)
A\a- f (u) Switch Stoi
?? simulation
non-zero
Mux2 when Rt-Rm
sgrt(u[4] A
2 Rt-Rm is negative
tu[5] 2+u[6)A2)
A

Figure 75. Proportional Navigation CPA and Stop Simulation

Calculation.

f (u) ♦ cpaactual
Missile A A CPA of Actual
sqrt((u[l]-u[4]) 2+(u[2]-u[5]) 2+ Target
and (u[3]-u[6]r2)
Target
Actual "Flight

Figure 76. Proportional Navigation CPA Calculation.

67
G. COMMAND GUIDANCE SIMULINK MODEL

©-+c time
Clock

!3> r^
M ^
Sigma Yaw
Calculation CRE_y
Target
Flight2
Range
Target

>
3> >
Sigma Pitch Mux
Calculation
CPA1 Mux2 CRE_p

a% Missile
Mux

Muxl
:^
CRE

Flight

3-
Sigma
Calculation

Figure 77. Command Guidance Missile Model

68
 r* i

position
MATIAB
Function ►I est |
target fibi_ estimated
theta Kaiman target position
l/s Filter
Vxt xt(0)

V*r x position
yt» target with
y position
noise
target
■2121 l/s
^ ^ -*\ yt ►B
y position Target
Vyt yt(0) observed
target with
Cartesian Spherical noise flight
to to
1000 Spherical Cartesian

\ z position
target with
position
target noise

Figure 78. Command Guidance Target Flight with Noise,

r-l xt |
X position
target
-2121 l/s
Vx Integrator

1—► yt
>osita on
U
y rtarget
-2121 l/s
Vy Integratorl Mux .171
•HI
Target
flight
1000
zt

zt |
H
z position
Mux 3

target

Figure 79. Command Guidance Target Flight without Noise,

69
 velocity
ax x 1-
f(u)
1 -^Demiö sin(u[l]) 7* accel x x position
missile
crey Demuxl
creydot
1XD+
Sum2 -966
1/s
i5 -3500
» 1/s
■0
966 3000
u[2])sin(u fc
iO
accel y velocity
sigy. y
lllüU—tit * 1 -1
IT is<u[in
cos(u[l])^V- L^g] »ppi 1/s 1/s
sigp 13
^ftu) -3500 il ym
3500 L
2])cos(u
[u[2])cos(u TTrV"J »r^mn
L+[7] y positioh
vy missile
az
rß>
1^ 1 accel 2
40
fc©
4 -^Demux
196
£<jö>
cos(u[2] )
i* -966
1/s
14
+ Z7 1/s -HU
-3500 i2
crep Demux
crepdot
966 3500 *0
z position
^¥\ vz 1 missile
velocity
z

Figure 80. Command Guidance Missile Flight,

70
 1 f (u) 1
/v /s A
target sqrt(xm 2+ym 2+zm 2) Rt
flight

Figure 81. Command Guidance Target Range,

sigmty
sigma yaw target
»f (u) sigma yaw missile
Target atan2(u[3],u[l])
Flight

Mux
f (u) sigma yaw target
atan2(u[3],u[l]) sigma yaw missile
Missile Mux2
Flight

Figure 82. Command Guidance Sigma Yaw Calculation.

71
 ► sigmtp
sigma pitch target
sigma pitch missile

flu)
target atan2(u[5],sqrt(u[l]~2+u[3]A2)) Mux
flight
sigma pitch target
sigma pitch missile
2
_ f (u)
Mux2
missile atan2(u[5],sqrt(u[l]~2+u[3]"2))
flight

Figure 83. Command Guidance Sigma Pitch Calculation.

sigy I
Demux ,
sigma
target yaw
Demuxl
f (u)
atan2((u[2]-u[6]),(u[l]-u[4]J)
►0
sigma
yaw
sigp 1
sigma
pitch
Mux -»If (u)h
Demux atan2((u[3]-u[B]), sigma
sqrt((u[l]-u[4})^2+(u[2]-u[6])A2) pi?ch
missile
Mux
Demux

Figure 84. Command Guidance Sigma Calculation.

72
 Demux
ores I
target
flight
_ Cross
Demuxl 1 1/u Range Error
Rt 1/Rt

WD- -a
Cross
Range Error
Mux f (u)
A
sqrt((ymzt-ytzm) 2+
(xtzm-xmzt)A 2+
Demux A
(xmyt-xtym) 2)
Mux
missile
flight
Demux

J
Figure 85. Command Guidance Cross Range Error.

-*] crey |
ere yaw

ere yaw
f (u)
missile sqrt(u[l]*2+u[3]A2)
flight

I CREl Mux
f (u) T » ere yaw
Sigma yaw target sin (u[l]-u[2]) Muxl credot yaw
sigma yaw missile

^du/dt^ »|creyd|
Derivative cross range
error dot
yaw

Figure 86. Command Guidance Cross Range Error Yaw,

73
 crep 1

1 fc MATLAB
Function
pitch target 3ign(u[l]-u[2])
sigma
pitch missile
,
trn
^ '^ ' 1
M •! cie piLch
Muxl credot pitch

2
_
f (u)
CRE sqrt(u [11-2-u 21*2)
UKt yaw
J►|o :i
Derivative CREdot pitch

Figure 87. Command Guidance Cross Range Error Pitch.

1
xt

2 f (u) * cpaactual
* sqrt((u[l]-u[4))*2 actual target CPA
yt
+ (u[2]-u[5])»2
+(u[3]-u[6])*2)
3 Mux
zt
-+■
♦ f (u)
4 Demux Mux sgrt(u[l]*2
Mi ssi le * +u[2]A2+u[3]A2) Constant
Flight Demux
*1ux-* f(u) —j*H STOP
Mux2 Rt-Rm Stop simulation
Switch i5 non-zero
when
f (u) rt-rm
is less than
+ufi ]A
2+u[6]-2) zero

Figure 88. Command Guidance CPA Calculation.

74
 1
xt -»tfux-
range
tfux- f (u) f (u) —►TT
2 ÜUX-.
+
yt sqrt(u[l]A2+
u[2]A2+u[3]A2)
[ti',U'] V p]li
r noise pili

f (u)
#2+ * 2
phi phi
[ti' ,W] theta
phi noise theta

f (u)
theta + theta

[ti',V]
theta noise

Figure 89. Cartesian to Spherical Block for both

Simulations.

75
H. MISCELLANEOUS MATLAB CODE

% This program generates the noise used in the target flight.

randn(•seed1,26579);
ti= [0: .001:30] ;
for i= 1:30001
%Range noise
U(i) = randn*l5;
%Pitch angle noise
V(i) = randn*pi/l80;
%Yaw angle Noise
W(i) = randn*pi/l80;
end
%This program sets the initial conditions for the Kaiman
%Filter. It is run at the beginning of each simulation,
clear P
clear xhat
global P
global xhat
%initial covariance matrix
P=le6*eye(6);
%initial estimated target position
xhat= [10000 -500 1000 -500 0 500]';
%This function runs a Kaiman filter algorithm
%for the given A, B, C matices for constant velocity flight
function[xhat,P]=klmn(u,P,xhat);
%initialization
A= [0 1 0 0 0 0
0 0 0 0 0 0
0 0 0 10 0
0 0 0 0 0 0
0 0 0 0 0 1

76
 0 0 0 0 0 0];
B=[0;0;0;0;0;0] ;
C= [1 0 0 0 0 0;
0 0 10 0 0;
0 0 0 0 10];
%Time step and q parameter for the Kaiman Filter
q=l;
dt=.001;
%Continuous to discrete conversion
[phi,del]=c2d(A,B,dt);
%Specify position and angle vectors
pos=[u(l) ;u(2) ;u(3)] ;
ang= [u(4) ;u(5) ;u(6)] ;
%Calculate Sigma Matrix
sigma= [ ((qA2)* (dt) *3)/3 ((q*2)*(dt) *2)/2;
((q*2)*(dt)*2)/2 (q'2)*(dt)];
%Calculate Q Matrix
0=[sigma zeros(2,2) zeros(2,2);
zeros(2,2) sigma zeros(2,2);
zeros(2,2) zeros(2,2) sigma];
%Calculate Rstar Matrix
Rstar= [225 0 0;0 (I*pi/180) "2 0;0 0 (I*pi/180) "2] ;
%Kalman iteration
xhatkpl = phi*xhat;
Pkpl = phi*P*phi' + Q;
H = calch(ang);
R = H*Rstar*H';
K = Pkpl*C'*inv(C*Pkpl*C'+R) ;
Pklkl = (eye(6)-K*C)*Pkpl*(eye(6)-K*C)' + K*R*K';
xklkl = xhatkpl + K*(pos(:,1)-(C*xhatkpl));
global xhat;

77
 xhat = xklkl;
global P;
P = Pklkl;
%This function calculates the H matrix fox computation
%in a Kaiman filter
function [H]=calch(x)
a=(tan(x(2)))"2;
b=(tan(x(3)))"2;
c=(sec(x(2)))'2;
d= (sec(x(3) ) ) "2;
e=tan(x(2) ) ;
f=tan(x(3));
H = [l/((a*b+a+b+l)".5) -x (1)* (b*e*c+c*e)/((a*b+a+b+l)*1.5)
-x(l)*(a*d*f+d*f)/((a*b+a+b+l)"1.5);
e/ ( (a*b + a + b + l) " .5)
-x(l) * (a*c* (b+1) ) / ( (a*b+a+b+l) *1.5)+x(l) *c/( (a*b+a+b+l) * .5)
-x(l)*e*f*d*(a+1)/((a*b+a+b+l)"1.5) ;
f/((l+b)*.5) o -x(l)*b*d/((i+b)*l.5)+x(l)*d/((i+b)*.5)] ;

78
 BIBLIOGRAPHY

Blackelock, J.H., Automatic Control of Airczaft and

Missiles, Wiley-Interscience Publishing, New York, NY. 1991.

Davis, H.F., Introduction to Vector Analysis, Wm.C. Brown

Publishing, Dubuque, IA. 1991.

Hostetter, G.H., Santana, M.S., Stubberud, A.R., Digital

Control System Design, Saunders College Publishing, Fort
Worth, TX. 1994.

Peppas, D.I, "A Computer Analysis of Proportional

Navigation and Command to Line of Sight of a Command Guided
Missile for a Point Defence System," M.S. Thesis, Naval
Postgraduate School, Monterey CA, December 1992.

Titus, H.A., Missile Guidance, unpublished notes, Naval

Postgraduate School, Monterey, CA.

79
80
 INITIAL DISTRIBUTION LIST

No. Copies
Defense Technical Information Center 2
Cameron Station
Alexandria, Virginia 22304-6145
Library, Code 52 2
Naval Postrgaduate School
Monterey, California 93943-5101
Chairman, Code EC 1
Department of Electrical and Computer Engineering
Naval Postgraduate School
Monterey, California 93943-5121
Professor H.A. Titus, Code EC/Ts 1
Department of Electrical and Computer Engineering
Naval Postgraduate School
Monterey, California 93943-5121
Professor R.G. Hutchins, Code EC/Hu 1
Department of Electrical and Computer Engineering
Naval Postgraduate School
Monterey, California 93943-5121
Professor P. Pace, Code EC/Pc 1
Department of Electrical and Computer Engineering
Naval Postgraduate School
Monterey, California 93943-5121
Lieutenant Patrick Costello 1
9 Rhonda Dr.
Scarborough, Maine 04 074

81