Planejamento de rota de pouso; Aplicação de VANTs; Algoritmos Evolutivos; Algoritmos Genéticos; Estratégias Heurísticas; Otimização.

1. A Multi-Population Genetic Algorithm for UAV
Path Re-Planning under Critical Situation
Jesimar S. Arantes (USP)
Márcio S. Arantes (USP)
Claudio F. M. Toledo (USP)
Brian C. Williams (MIT)
São Carlos, SP
November – 2015
Jesimar S. Arantes (USP)Márcio S. Arantes (USP)Claudio F. M. Toledo (USP)Brian C. Williams (MIT) (USP)IEEE ICTAI 2015 November – 2015 1 / 41

2. Outline
1 Introduction
2 Problem Description
3 Methods
4 Computational Results
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 2 / 41

3. Introduction
Overview
Figure 1: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 3 / 41

4. Introduction
Overview
Figure 2: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 4 / 41

5. Introduction
Overview
Figure 3: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 5 / 41

6. Introduction
Overview
Figure 4: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 6 / 41

7. Introduction
Overview
Figure 5: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 7 / 41

8. Introduction
Overview
Figure 6: Illustrative scenario for mission planning.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 8 / 41

9. Problem Description
Types of Regions and Critical Situation
Regions
1 No-Fly Zone (φn)
2 Penalty Region (φp)
3 Bonus Region (φb)
4 Remainder Region (φr )
Critical Situation
1 Motor Failure (ψm)
2 Battery Failure (ψb)
3 Aerodynamic Surfaces Failure
type 1 (ψs1 )
4 Aerodynamic Surfaces Failure
type 2 (ψs2 )
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 9 / 41

10. Problem Description
Types of Regions and Critical Situation
Regions
1 No-Fly Zone (φn)
2 Penalty Region (φp)
3 Bonus Region (φb)
4 Remainder Region (φr )
Critical Situation
1 Motor Failure (ψm)
2 Battery Failure (ψb)
3 Aerodynamic Surfaces Failure
type 1 (ψs1 )
4 Aerodynamic Surfaces Failure
type 2 (ψs2 )
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 9 / 41

11. Methods
Codification, Decodification and Solution
Codification ut:
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 10 / 41

12. Methods
Codification, Decodification and Solution
Codification ut:
Decodification FΨ:
xt+1 = FΨ(xt , ut )


px
t+1
py
t+1
vt+1
αt+1

 =


px
t + vt · cos(αt ) · ∆T + at · cos(αt ) · (∆T)2/2
py
t + vt · sen(αt ) · ∆T + at · sen(αt ) · (∆T)2/2
vt + at · ∆T − Fd
t
αt + εt · ∆T


Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 10 / 41

13. Methods
Codification, Decodification and Solution
Codification ut:
Decodification FΨ:
xt+1 = FΨ(xt , ut )


px
t+1
py
t+1
vt+1
αt+1

 =


px
t + vt · cos(αt ) · ∆T + at · cos(αt ) · (∆T)2/2
py
t + vt · sen(αt ) · ∆T + at · sen(αt ) · (∆T)2/2
vt + at · ∆T − Fd
t
αt + εt · ∆T


Solution xt:
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 10 / 41

14. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 11 / 41

15. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 12 / 41

16. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 13 / 41

17. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 14 / 41

18. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 15 / 41

19. Methods
Greedy Heuristic
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 16 / 41

20. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 17 / 41

21. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 18 / 41

22. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 19 / 41

23. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 20 / 41

24. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 21 / 41

25. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 22 / 41

26. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 23 / 41

27. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 24 / 41

28. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 25 / 41

29. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 26 / 41

30. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 27 / 41

31. Methods
Multi-Population Genetic Algorithm
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 28 / 41

32. Methods
Objective Function
minimize fitness = −Cφb
·
|φb|
i=1
(P(xK ∈ Zi
φb
)) + Cφp ·
|φp|
i=1
(P(xK ∈ Zi
φp
)) +
Cφn · max(0, 1 − ∆ − P
K
t=0
|φn|
i=1
xt /∈ Zi
φn
) + 1
|εmax |
·
K
t=0
ut · |εt | +
shortestDist(xK , Zφb
) +
Cφb
, vK − vmin > 0
0 , otherwise
+ Cφb
· 2
(K−T)
10 , ψ = ψb
0 , otherwise
Landing on φb
Landing on φp
Landing and fly on φn
Curves of the UAV
Distance to φb
Time violation
Battery failure
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 29 / 41

33. Methods
Methods Used
In this work, the following methods were used.
GH: Greedy Heuristic
MPGA1(–GH): Multi-Population Genetic Algorithm 1
Without greedy operator
MPGA2(+GH): Multi-Population Genetic Algorithm 2
With greedy operator
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 30 / 41

34. Computational Results
Automatically Generated Maps
Level of Difficulty
1 ME : a), b)
2 MN: c), d)
3 MH: e), f)
Level of Coverage
1 C25%: a), c), e)
2 C50%: b), d), f)
Legend Colors
1 φb
2 φp
3 φn
4 φr
(a) (b) (c)
(d) (e) (f)
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 31 / 41

35. Computational Results
Parameters and Settings used in the Experiments
Model Parameters Value
Map
Dimension X [m] 1000
Dimension Y [m] 1000
UAV
Initial Position (px
0 , py
0 ) [m] (0; 0)
Initial Velocity (v0) [m/s] 24
Initial Angle (α0) [o
] 90
Linear Velocity (vmin; vmax ) [m/s] [11; 30]
Angular Variation (εmin; εmax ) [o
/s] [−3; 3]
Acceleration (amin; amax ) [m/s2
] [0; 2]
Number of time steps to land (T) [s] 60
Time Discretization (∆T) [s] 1
Probability of failure (∆) 0.001
MPGA
Populations 3
Individuals/Pop 13
Individuals Total 39
Mutation Rate 0.5
Crossover Rate 0.75
Stop Criterion 10000
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 32 / 41

36. Computational Results
Experiments: Result obtained for the GH, MPGA1(–GH) and MPGA2(+GH)
GH MPGA1(–GH) MPGA2(+GH)
Ψ Instance φb φr Inf. φb φr Inf. φb φr Inf.
ψm
ME and C25% 79 21 0 81 19 0 90 10 0
ME and C50% 92 6 2 92 7 1 96 3 1
MN and C25% 58 39 3 60 39 1 71 28 1
MN and C50% 86 12 2 84 16 0 96 4 0
MH and C25% 30 52 18 36 64 0 40 60 0
MH and C50% 62 28 10 60 33 7 82 15 3
Avg 67.8 26.3 5.8 68.8 29.7 1.5 79.2 20.00 0.83
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 33 / 41

37. Computational Results
Experiments: Result obtained for the GH, MPGA1(–GH) and MPGA2(+GH)
GH MPGA1(–GH) MPGA2(+GH)
Ψ Instance φb φr Inf. φb φr Inf. φb φr Inf.
ψb
ME and C25% 99 0 1 100 0 0 100 0 0
ME and C50% 97 0 3 99 0 1 99 0 1
MN and C25% 93 3 4 94 5 1 99 0 1
MN and C50% 98 0 2 99 0 1 100 0 0
MH and C25% 67 5 28 73 27 0 94 6 0
MH and C50% 83 0 17 68 17 15 95 2 3
Avg 89.5 1.3 9.2 88.8 8.2 3.0 97.8 1.3 0.8
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 34 / 41

38. Computational Results
Experiments: Result obtained for the GH, MPGA1(–GH) and MPGA2(+GH)
GH MPGA1(–GH) MPGA2(+GH)
Ψ Instance φb φr Inf. φb φr Inf. φb φr Inf.
ψs1
ME and C25% 81 8 11 90 8 2 91 7 2
ME and C50% 88 0 12 89 0 11 93 0 7
MN and C25% 68 16 16 76 18 6 86 8 6
MN and C50% 82 1 17 84 3 13 89 0 11
MH and C25% 41 23 36 49 46 5 67 28 5
MH and C50% 56 0 44 46 23 31 78 4 18
Avg 69.3 8.0 22.7 72.3 16.3 11.3 84.0 7.8 8.2
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 35 / 41

39. Computational Results
Experiments: Result obtained for the GH, MPGA1(–GH) and MPGA2(+GH)
GH MPGA1(–GH) MPGA2(+GH)
Ψ Instance φb φr Inf. φb φr Inf. φb φr Inf.
ψs2
ME and C25% 90 4 6 94 4 2 99 0 1
ME and C50% 90 0 10 95 1 4 95 1 4
MN and C25% 70 20 10 79 16 5 92 5 3
MN and C50% 87 1 12 83 8 9 94 0 6
MH and C25% 40 17 43 62 35 3 74 24 2
MH and C50% 61 3 36 57 13 30 76 4 20
Avg 73.0 7.5 19.5 78.3 12.8 8.8 88.3 5.7 6.0
Avg Final 74.9 10.8 14.3 77.1 16.7 6.2 87.3 8.7 4.0
Time (Sec)
GH MPGA1 MPGA2
0.07 1.017 0.874
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 36 / 41

40. Computational Results
Experiments: Example of Routes
S
E
(a)
S
E
(b)
Figure 7: Routes determined by the planner MPGA2(+GH) in a map MN with
coverage C25%: (a) ψm. (b) ψb.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 37 / 41

41. Computational Results
Experiments: Example of Routes
S
E
(c)
S
E
(d)
Figure 8: Routes determined by the planner MPGA2(+GH) in a map MN with
coverage C25%: (c) ψs1 . (d) ψs2 .
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 38 / 41

42. Computational Results
Experiments: Example of Routes
(a) Wind Velocity 10 Knots - MN with C25% (b) Wind Velocity 50 Knots - MN with C25%
Figure 9: (a), (b) FG simulation with winds 10 and 50 knots. Wind direction:
west.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 39 / 41

43. Computational Results
Video FlightGear Simulator
Figure 10: Video FlightGear Simulator.
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 40 / 41

44. Acknowledgements
Questions send email to:
jesimar.arantes@usp.br
marcio, claudio@icmc.usp.br
williams@mit.edu
Thank You!
Jesimar S. Arantes (USP) IEEE ICTAI 2015 November – 2015 41 / 41