Gregory J. Barlow

1

Design of Autonomous Navigation Controllers for
Unmanned Aerial Vehicles using Multi-objective
Genetic Programming

• Gregory J. Barlow
• March 19, 2004

2
Overview

• Background
• Unmanned Aerial Vehicle Control
• Evolution and Fitness Evaluation
• Experiments and Results
• Conclusions and Future Work

3
Evolutionary Computation

• Biologically inspired computational method of
  problem solving
• May be applied to a variety of structures (binary
  strings, real numbers, computer programs,
  hardware, neural networks, etc) because the
  algorithm operates on an encoding of the
  parameters, not the parameters themselves

4
Genetic Programming

• A population of random programs is created
• Each individual in the population undergoes a
  fitness test and is assigned a fitness value
• Genetic operators (crossover, mutation, etc) are
  performed on the population to form the next
  generation
• The process is repeated until a suitable
  individual is evolved

5
Evolutionary Process
6
Representation

• Each individual is a program, which we represent
  as a tree
• Function set for non-leaf nodes
• Terminal set for leaf nodes

7
Crossover
8
Mutation
9
Unmanned Aerial Vehicle Control

• Create controllers that will fly a UAV toward a
  target radar and then circle the radar for
  jamming
• Make the UAV controller completely autonomous
• Be able to handle multiple radar types
• Be able to transfer evolve controllers to real
  UAVs

10
Simulation

• To test the fitness of a controller, the UAV is
  simulated for 4 hours of flight time in a 100 by
  100 square nmi area
• The initial starting position of the UAV is
  randomly set along the bottom of the simulation
  space
• The position of the radar is also randomly set
  for each simulation
• UAVs can sense the AoA and amplitude of incoming
  radar signals

11
Simulation
12
Transference

• These controllers should be transferable to real
  UAVS. To encourage this
• Only the sidelobes of the radar were modeled
• Noise is added to the modeled radar emissions
• The angle of arrival value from the sensor is
  only accurate within 10

13
Functions and Terminals

• Hard Left, Hard Right, Shallow Left, Shallow
  Right, Wings Level, No Change
• IfThen, IfThenElse, And, Or, Not, lt, lt, gt, gt, gt
  0, lt 0, , , -, , /
• Amplitude gt 0, Amplitude Slope lt 0, Amplitude
  Slope gt 0, AoA lt, AoA gt

14
Fitness Functions

• Normalized distance
• Circling distance
• Level time
• Turn cost

15
Normalized Distance
16
Circling Distance
17
Level Time
18
Turn Cost
19
Performance of Evolution

• Multi-objective genetic programming produces a
  Pareto-optimal front of solutions, not a single
  best solution.
• To gauge the performance of evolution, fitness
  values for each fitness measure were selected for
  a minimally successful controller.

20
Baseline Values

• Normalized Distance 0.15
• Circling Distance 4
• Level Time 1000
• Turn Cost 0.05

21
Direct Evolution Experiments

• Continuously emitting, stationary radar
• Intermittently emitting, stationary radar with a
  regular period
• Intermittently emitting, stationary radar with an
  irregular period
• Continuously emitting, mobile radar
• Intermittently emitting, mobile radar with a
  regular period

22
Direct Evolution
Radar Type Runs Runs Runs Controllers Controllers Controllers
Radar Type Total Succ. Rate Total Avg. Max.
Continuous, Stationary 50 45 90 3,149 62.98 170
Intermittent, stationary (regular period) 50 25 50 1,891 37.82 156
Intermittent, stationary (irregular period) 50 29 58 2,374 47.48 172
Continuous, mobile [phone]
Intermittent, mobile (regular period) 50 16 32 569 11.38 93
23
Continuously emitting, stationary radar
24
Circling Behavior
25
Intermittently emitting, stationary (regular)
26
Intermittently emitting, stationary (irregular)
27
Continuously emitting, mobile radar
28
Intermittently emitting, mobile radar
29
Incremental Evolution

• Continuously emitting, stationary radar (seed
  populations)
• Intermittently emitting, stationary radar
• Continuously emitting, mobile radar
• Intermittently emitting, stationary radar
  (multiple increments)
• Intermittently emitting, mobile radar (multiple
  increments)

30
Incremental Evolution
Radar Type Runs Runs Runs Controllers Controllers Controllers
Radar Type Total Succ. Rate Total Avg. Max.
Continuous, Stationary 50 45 90 2,815 56.30 166
Intermittent, stationary 50 34 64 2,526 50.52 184
Continuous, mobile [phone]
Intermittent, stationary (multiple increments) 50 42 84 2,083 41.66 143
Intermittent, mobile (multiple increments) 50 37 74 1,602 32.04 143
31
Intermittent, mobile (multiple increments)
32
Transference to a wheeled mobile robot

• Controllers were designed for UAVs
• A UAV was not yet available for flight tests to
  evaluate transference
• Evolved controllers were tested on a wheeled
  mobile robot, the EvBot II
• A speaker was used in place of the radar, and an
  acoustic array in place of the radar sensor

33
EvBot II

• PC/104 processor
• Communications with a wireless network card
• Runs Linux
• On-board acoustic array

34
Transference considerations

• In simulation, the sensor accuracy was 10, but
  the acoustic array accuracy was approximately
  45
• Wheeled robot not controlled by roll angle, must
  be turned and then moved
• The size of the maze environment was not
  equivalent to the simulation environment, instead
  the scale size of the maze environment was 1.13
  by 0.9 nautical miles

35
Sensor accuracy

• Sensor accuracy of 10 Sensor accuracy of 45

36
Controller 1
37
Controller 2
38
Conclusions

• Autonomous navigation controllers were evolved to
  fly to a radar and then circle around it while
  maintaining stable and efficient flight dynamics
• Multi-objective genetic programming was used to
  evolve controllers
• Controllers were evolved for five radar types
  using both direct evolution and incremental
  evolution

39
Conclusions

• Incremental evolution dramatically increased the
  success rates for the more difficult radar types
• Methods were used to aid in transference of
  controllers to real UAVs
• Controllers were tested on a wheeled mobile robot
  with good success
• Evolved controllers are capable of transference
  to real physical vehicles

40
Future Work

• Controllers will be tested on physical UAVs for
  several radar types
• Distributed multi-agent controllers will be
  evolved to handle cases of multiple UAVs against
  multiple radars
• Incremental evolution will be used to aid in the
  evolution of fit multi-agent controllers for
  complex radar types