Optimal Path Planning and Power Allocation for a Long Endurance Solar-Powered UAV

Optimal Path Planning and Power Allocation for a Long Endurance Solar-Powered

UAVSaghar Hosseini, Ran Dai, and Mehran Mesbahi

1

Saghar Hosseini, ACC 2013, Washington, DC

Robotics, Aerospace, and Information Networks Lab

Motivation and Applications• Development of Solar-Powered UAVs

• The first solar-powered flying models was built in 1974

• NASA pathfinder was launched in 1998 with altitude of 80,201ft

• Boeing Solar Eagle is expected to stay aloft for five years• Applications :

2

Atmospheric SatelliteSurveillance and Reconnaissance

Forest Fire Fighting

Saghar Hosseini, ACC 2013, Washington, DC

Outline• Previous work• Optimal control problem formulation• Nonlinear programming results• Reduced hybrid model• Examples• Future research Th

e 20

13 A

mer

ican

Con

trol

Co

nfer

ence

, Jun

e 17

- 19

, W

ashi

ngto

n, D

C

3

Saghar Hosseini, ACC 2013, Washington, DC

Previous Work• Level flight

• Klesh and Kabamba (2007, 2009)

• Cylinder• Spangelo et al. (2009)

• 3D trajectory• Sachs et al. (2009)

4

Saghar Hosseini, ACC 2013, Washington, DC

Solar UAV Power Resources

5

Saghar Hosseini, ACC 2013, Washington, DC

Solar UAV Kinematics

6

Saghar Hosseini, ACC 2013, Washington, DC

Optimal Control Problem

7

Saghar Hosseini, ACC 2013, Washington, DC

Nonlinear Programming Results

8

-0.5 0 0.5 1 1.5 2-1

-0.5

0

0.5

1

1.5

2

y (1

00km

)

x (100km)

0 5 10 15 20 250

1

2

3

4

5

6

7

8

z (k

m)

time (hr)

0 5 10 15 20 25-0.5

0

0.5

(d

eg)

time (hr)

0 5 10 15 20 25-3

-2

-1

0

1

2

3

4

(de

g)

time (hr)

Saghar Hosseini, ACC 2013, Washington, DC

Level flight

Climb Glide

Level flight

Climb Glide

Nonlinear Programming Results

9

0 5 10 15 20 25-40

-20

0

20

40

60

80

100

time (hr)

Pow

er (w

att)

PBattPSunPEng

0 5 10 15 20 25

0.4

0.5

0.6

0.7

0.8

0.9

1

SO

C

time (hr)

Saghar Hosseini, ACC 2013, Washington, DC

Level flight

Climb Glide

Level flight

Climb Glide

0 5 10 15 20 250

5

10

15

V (m

/s)

time (hr)

Nonlinear Programing Results

10

Saghar Hosseini, ACC 2013, Washington, DC

Reduced Hybrid Model

11

Saghar Hosseini, ACC 2013, Washington, DC

Mode 1 : Low Altitude Level Flight

12

Saghar Hosseini, ACC 2013, Washington, DC

Mode 2 : Steady climb

13

Saghar Hosseini, ACC 2013, Washington, DC

Examples

14

-2 -1 0 1 2-1

-0.5

0

0.5

1

1.5

2

y (1

00km

)

x (100km)

Reduced ModelOriginal model

0 5 10 15 20 250

1

2

3

4

5

6

7

8

z (k

m)

time (hr)

Reduced ModelOriginal model

0 5 10 15 20 25-3

-2

-1

0

1

2

3

4Flight path angle (deg)

(de

g)

time (hr)

Reduced ModelOriginal model

0 5 10 15 20 25-0.5

0

0.5

(d

eg)

time (hr)

Reduced ModelOriginal model

Saghar Hosseini, ACC 2013, Washington, DC

Examples

15

Original Model

Reduced Hybrid Model

393.417 sec 1.758 sec

0 5 10 15 20 25

0.4

0.5

0.6

0.7

0.8

0.9

1

SO

C

time (hr)

Reduced ModelOriginal model

0 5 10 15 20 25-40

-20

0

20

40

60

80

100

time (hr)

Pow

er (w

att)

PBatt

PSun

PEngPBatt Original model

PSun Original model

PEng Original model

Saghar Hosseini, ACC 2013, Washington, DC

0 5 10 15 20 250

5

10

15

V (m

/s)

time (hr)

Reduced ModelOriginal model

Computation time

Reduced Hybrid Model Result

16

Saghar Hosseini, ACC 2013, Washington, DC

Future Research

17

• Coverage Problem• Thermal fields/Wind gust• Coordinated flight• Mission planning

Saghar Hosseini, ACC 2013, Washington, DC