Date post: | 24-Dec-2015 |
Category: |
Documents |
Upload: | kathryn-jackson |
View: | 220 times |
Download: | 0 times |
1DAT/SENMar 22, 2006
The Junior Woodchucks Guidebook
for Fighter Pilots
Lars Rosenberg RandleffPh.D. student
Danish Defence Research Establishment
2DAT/SENMar 22, 2006
Junior Woodchucks Guidebook
• In Disney's fictional Scrooge McDuck universe, The Junior Woodchucks are the Boy Scout-like youth organization to which Donald Duck's nephews, Huey, Dewey and Louie, belong.
Source: Wikipedia.org
• The nephews always carry with them a copy of the Junior Woodchucks Guidebook, a fictional guidebook which is full of detailed and pertinent information about whatever country or situation in which Donald and the boys find themselves. Its depth of coverage is remarkable, considering that it is a small paperback book.
3DAT/SENMar 22, 2006
Vision for the project
• Numbers of threats, guidance methods and countermeasures are increasing• 600 missile systems developed• 200-300 missile systems currently deployed
(http://missile.index.ne.jp)
• Not easy to find the best action• Takes training to do it in time• Solution depends on mission
• “The worlds best pilot as the co-pilot”
4DAT/SENMar 22, 2006
Missile guidance
Semi-active Radar
Infrared
Electro-Optical
Active Radar
Track-Via-Missile
INS/GPS
Command
Beam Rider
Anti-Radiation
5DAT/SENMar 22, 2006
Data flow
Position
DSS
Situational Awareness
Decision Presentation to the pilot
• Audio?• Display?• Force feedback?
RWR
MWS
Velocity
Orientation
Jammer
Chaff
Flares
Mission data
Link-16 data
Knowledge Base
• AA• SA• Single threat• Multiple threats
Jammer
Chaff
Flares
6DAT/SENMar 22, 2006
Data flow
Position
DSS
Situational Awareness
Decision Presentation to the pilot
• Audio?• Display?• Force feedback?
RWR
MWS
Velocity
Orientation
Jammer
Chaff
Flares
Mission data
Link-16 data
Knowledge Base
• AA• SA• Single threat• Multiple threats
Jammer
Chaff
Flares
7DAT/SENMar 22, 2006
Approaches
• AI techniques• Bayesian Networks• Prolog-based Expert
System• Artificial Neural
Networks
• OR techniques• Mathematical
Modelling• Metaheuristics
9DAT/SENMar 22, 2006
Dependency tables
• A lot of work to fill them• Not easy without firm knowledge of
all dependencies
• Can be done using HUGIN (Structural Learning)• Not easy to get data
10DAT/SENMar 22, 2006
Using Fly-In
• Uses models to simulate an IR guided missile approach
• Can generate data to be used in Structural Learning
• Has a lot of parameters!
11DAT/SENMar 22, 2006
Looking ahead
• Decisions depend on what to expect
• Maximizing the survivability• Lethality vs. Survivability• What to optimize?
• Defining scenarios• Threats• Countermeasures
12DAT/SENMar 22, 2006
Survivability
tot
iii
iiii
N
iiitot
LS
RRR
MTRLTPL
1
)),,,(1()(
,,
1
n
ttprod
tt
best
n
ttsum
SS
SS
SS
0
0
}{max
13DAT/SENMar 22, 2006
Efficiency of CM (α)
0.2
0.4
0.6
0.8
1
30
210
60
240
90
270
120
300
150
330
180 0
r = getEffAlpha(t, chaff)
0.2
0.4
0.6
0.8
1
30
210
60
240
90
270
120
300
150
330
180 0
r = getEffAlpha(t, jammer)
0.2
0.4
0.6
0.8
1
30
210
60
240
90
270
120
300
150
330
180 0
r = getEffAlpha(t, decoy)
14DAT/SENMar 22, 2006
Threat Lethality (II)
-1 0 1 2 3 4 5
x 104
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4x 10
4
-1 0 1 2 3 4 5
x 104
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4x 10
4
0 50 100 150 200 250 300 350 400 450 5000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Scenario Lethality