Tutorial ASTRODYNAMICS PDF

Fundamentals of Astrodynamics
and Applications
By David A. Vallado
Tutorial Lectures at the
4th ICATT, Madrid, Spain
April 30, 2010
May 3-6, 2010
Pg 1 of 120 www.centerforspace.com
Objectives
• Use an example problem to illustrate various
astrodynamic techniques you’ll need to know
• Introduce you to the various topics that the text

covers in more detail
Problem Scenario
• Determine when you can see a satellite from
a ground site
• What we’ll need to understand

– Time
– Coordinate systems
– Propagation
– Orbit Determination
– ... and some others ☺
What we’ll cover
• Fundamental Concepts
– Time and Coordinate Systems
• Newton
– Equations of Motion
• Kepler
– Equation
– Problem
– Satellite state
• Perturbations/Propagation
– Special
– General
• Orbit Determination and Estimation
• Applications
Fundamentals of Astrodynamics
and Applications Third Edition
Space Technology Library (Vol 21), Microcosm
Press/Springer
By David A. Vallado
Center for Space Standards and innovation
Paperback (ISBN 978-1-881883-14-2)
Published Spring 2007
http://astrobooks.com/index.asp?PageAction=VIEWPROD&ProdID=1137
US$ 60.00
My Objective with the Book
• Cover
– Fundamentals
– Some advanced material
• Bridge the gap in between
• Details
– Consistent notation
Chapter 3
• Newton
• Kepler
• Perturbations
• Orbit Determination
• Applications
Time and Coordinate Systems
• Essential, but not terribly exciting
What Time is it?
• 14:28
– Ok – That specifies that it’s afternoon
– But what time zone?
• Mountain Time is 6/7 hours before UTC (Greenwich,
Zulu)
– Need to specify
» Daylight Savings
» Standard Time
– Is that all? … No!
• TAI, TT (TDT), TDB, TCB, TCG, GPS, …
Solar and Sidereal Time
Earth Sun
Sidereal day
(23h 56m 4.0905s)
Stars
Reference Direction
Solar day
(24h)
Greenwich and Local Times
Local Meridian
VLST
l
LHAstar
Star
0°
VGMST
~ GHA~
Hour Angles vs Time?
• 24 hrs = 360 degrees
– Sidereal time assumed
Hour Angles
Local Observer
VLST
0°
VGMST
GHAstar
LHAstar East
astar
Star
~
What Time is it? (continued)
• Additional times
– UT1 (Universal Time, sidereal time)
• Solution from observations
• Shows slowly decreasing Earth rotation rate
– UTC is Coordinated Universal Time (solar time)
• “Clock time”
• Maintained within 0.9 s of UT1
– Leap Seconds
• UTC = UT1 + ΔUT1
– ΔUT1
» EOP Parameter that accounts for actual Earth rotation
» Calculated by USNO/ IERS
Time Scales
50.0
TCB
40.0
30.0
TCG TDB
Difference in Time to TAI (sec)
20.0
10.0
TAI
0.0
-10.0
UT1 GPS
-20.0
UTC
-30.0
-40.0
Jan-61 Jan-65 Jan-69 Jan-73 Jan-77 Jan-81 Jan-85 Jan-89 Jan-93 Jan-97 Jan-01 Jan-05 Jan-09 Jan-13
Summary for time
• Time
– Can be off by up to a second if no ΔUT1
– TT can be off a minute
• Used for many calculations
– Impact
• Seems small but …
– Consider satellite traveling at 7 km/s
– Many conversions necessary
• Satellite moves wrt sidereal time
• Clocks record Solar time
Coordinate Systems
• Sun based
– Heliocentric
– Barycentric
• Earth Based
– Geocentric (Inertial and fixed)
– Topocentric (fixed)
• Satellite Orbit Based
– Perficoal
– Radial vs Normal
– Equinoctial
• Satellite Based
– Attitude
Heliocentric Coordinate System
^
Z
Vernal equinox
1st day of spring
Summer solstice
~ Mar 21
1st day of Summer
~ Jun 21
Sun
Perihelion ~ 1 Jan
Aphelion ~ 1 Jul
Ecliptic Plane ^
Y
^ ~
X, Autumnal Equinox Vernal Equinox
1st day of Fall 1st day of winter
~ Sep 23 ~ Dec 21
Geocentric and Ecliptic
Coordinates
Geocentric Coordinate System
^
K
Equatorial Plane
^
I, ~ J^
Local Coordinate System
Orbit Based Systems - Perifocal
^
Q
^
W ^
K
^
P
Perigee, closest point to Earth
I^ J^
Orbit Based Systems – Normal and Radial
^
K
^ , cross-track
W
^ v ^, along-track
S
W
^ , radial
^
N I^ R
J^
v, T^, in-track
Orbit Based Systems - Equinoctial
Angular Measurements
• Latitude and longitude
– Familiar
• Right Ascension-Declination
– Optical measurements
Right Ascension - Declination
Motion of the Coordinate System
• Earth’s orbit is not exactly stable

– Precession
• Long period movement (~26000 years)
– Nutation
• Short period movement (~18.6 years)
• Fixed vs Inertial
– Sidereal Time
• Polar Motion
– Axis of rotation moves slightly over time
Precession and Nutation
Luni-solar precession effect
Nutation effect
Ecliptic plane
Planetary effect
Earth’s equator
Precession of Equinox Earth’s orbit

~ About Sun
Polar Motion
Equinox based
Celestial Reference Frame
IAU 1976 (zA ,θA, ζA)

MOD
IAU 1982 (∆ε, ∆ψ , ε0 , δ∆ε, δ ∆ψ ) ∆ε, ∆ψ

(Tables)
Precession [PN]
Nutation
TOD
Sidereal Rotation R3[θGAST · 1982]
PEF
Polar Motion R2[-xp] R1[-yp]
ITRF
Terrestrial Reference Frame

Traditional Traditional
1984 Procedures Interpolation
Equinox based CIO based
MHB2000 (zA ,θA, ζA) MHB2000 (χA, ωA, ψA, ε0) X, Y X, Y

MOD [BPN] (Series)
MHB2000 (∆ε, ∆ψ , ε0) X, Y, s

s (Series)
(Tables)
Bias-Precession [BPN] Eq. (xxx) = f ( X, Y, s )

Nutation
ERS CIRS
Sidereal Rotation R3[θGAST · 2000 ] R3[θERA ]
TIRS TIRS
Polar Motion R1[-yp] R2[-xp] R1[-yp] R2[-xp] R3[s’ ]
ITRF ITRF

Traditional Canonical 4-term Non Series Traditional
2003 Procedures Rotation Rotating Origin Interpolation
Equinox based CIO based
P03 (zA ,θA, ζA) P03 (χA, ωA, ψA, ε0) P03 (εA, ψJ, φJ, γJ) X, Y X, Y
MOD [BPN] (Series)
MHB2000 (∆ε, ∆ψ , ε0) X, Y, s

s (Series) or EO
+ optional 2006 rate adjustments (Tables)
Bias-Precession [BPN] Eq. (xxx) = f ( X, Y, s )

Nutation
ERS CIRS
Sidereal Rotation R3[θGAST · 2006 ] R3[θERA ]
TIRS TIRS
Polar Motion R1[-yp] R2[-xp] R1[-yp] R2[-xp] R3[s’ ]
ITRF ITRF

Traditional Canonical 4-term Fukushima – Non Series Traditional
Rotation Williams Rotating Origin Interpolation
2006 Procedures
Earth’s shape
• Oblate Spheroid
– An ellipsoidal approximation
• Other terms
– Geoids
• Gravity acts equally at all points on this surface
– Plumb-bobs will hang perpendicular
– Geopotential
• Mathematical representation of the precise
gravitational effect
Earth Surface
Earth Ellipsoid
• Convert
geocentric
(φgc) and
geodetic
(φgd) latitude
Chapter 1
• Newton
• Kepler
• Perturbations
• Applications
Newton’s Laws
• 1. Every body continues in its state of rest, or of uniform motion in a
right [straight] line, unless it is compelled to change that state by forces
impressed upon it.
• 2. The change of motion is proportional to the motive force impressed
and is made in the direction of the right line in which that force is
impressed.
• 3. To every action there is always opposed an equal reaction: or, the
mutual actions of two bodies upon each other are always equal and
directed to contrary parts. (Newton [1687] 1962, 13)
– The third law in Newton’s own words:
• If a horse draws a stone tied to a rope, the horse (if I may say so) will be
equally drawn back towards the stone; for the distended rope, by the same
endeavor to relax or unbend itself, will draw the horse as much towards the
stone as it does the stone towards the horse, and will obstruct the progress of
the one as much as it advances that of the other. (Newton [1687] 1962, 14)
Gravitational Law
• Forms the basis of Two-
body dynamics
– G is constant of gravitation = K K
Gm⊕ msat r
6.673x10-20 km3/kgs2 f gravity = − K
r2 r
Two-body Equation of Motion
• Simple form resulting from
K
rK = − G(m⊕ + msat ) rK
r2 r
• Newton
Chapter 2 • Kepler
• Perturbations
• Applications
Kepler’s Laws
• 1. The orbit of each planet is an ellipse with
the Sun at one focus.
• 2. The line joining the planet to the Sun
sweeps out equal areas in equal times.
• 3. The square of the period of a planet is
proportional to the cube of its mean distance
to the Sun.
Conic Sections
• All orbits
follow
– Circle
– Ellipse Ellipse
Hyperbola
Point
– Parabola Parabola
– Hyperbola
– Rectilinear
Circle
Nomenclature
• Kepler’s Equation and Kepler’s Problem
– Very different!
– Kepler’s equation
• Found during Kepler’s analysis of the orbit of Mars
– Kepler’s problem
• Generically used for propagating a satellite forward
– Usually two-body dynamics
Kepler’s Equation
• Find Eccentric
anomaly (E)
– E = 0º at ν = 0º,
180º
Kepler’s Problem
• Find future position

and velocity
– Given starting state
– Called propagation
Satellite State Representations
• Convey location of a satellite in space and time
• Types
– Numerical
• Position and velocity vectors
– Analytical (Elements)
• Classical (Keplerian, Osculating, two-body) (a, e, i, Ω, ω, ν)
• Equinoctial (af, ag, L, n, χ, ψ)
• Flight (λ, φgc, φfpa, β, r, v)
• Spherical (α, δ, φfpa, β, r, v)
• Canonical
– Delaunay
– Poincare
• Mean elements (theory dependant)
– Two-line element sets
» AFSPC, SGP4 derived, ‘mean’ elements
– ASAP
– LOP
– Other
• Other
– Semianalytical
• Theory dependant
Classical Orbital Elements
K v
Angular momentum, h
r
i υ Perigee, e
Equatorial Plane ω
J
I
Ω
Line of nodes, n
• Newton
• Kepler
Chapter 8/9 • Perturbations
• Applications
Introduction
• Several forces affect satellite orbits
– Gravitational
– Atmospheric Drag
– Third Body
• Sun, Moon, planets
– Solar Radiation Pressure
– Tides
• Solid Earth, Ocean, pole, etc.
– Albedo
– Thrusting
– Other
Applicability
Central Body Gravity
Atmospheric Drag
Third Body Gravity
Solar Radiation Pressure
Solid Earth Tides
Ocean Tides
Albedo
Other
100 1000 10,00 100,000

0
Orbital Altitude (km)
Perturbations
Mean Change Short-periodic plus long-periodic, and secular
c
Secular
Mean Change
Long-periodic and
secular
t1 t2 t3 t4
Time
Central Body Gravitational Forces
• Largest single contributor to the motion

– It’s why satellites stay in orbit!
• Conservative force
– Total kinetic and potential energy remains the same
GM ⎡⎢ ⎤
n
∞ n ⎛a⎞
V= 1+ ∑ ∑ ⎜ ⎟ nm P (sin φ )(C nm cos m λE + S nm sin m λE ⎥
)
r ⎢ ⎝ r ⎠ ⎥
⎣ n=2 m=0 ⎦
dm
1
( )
n
Pn (sin φ ) = sin 2 φ - 1 1 1
2n n! d (sin φ )
m
⎡ ( 2n + 1) k ( n − m)!⎤ 2 ⎧C ⎫ ⎡ ( n + m)! ⎤ 2 ⎧C nm ⎫
P =⎢ ⎥ P , and ⎨ nm ⎬ = ⎢ ⎥ ⎨ ⎬,
dm nm
⎣ ( n + m)! ⎦ nm ⎩ S nm ⎭ ⎣ ( 2n + 1) k ( n − m)!⎦ ⎩ S nm ⎭
Pnm (sin φ ) = (cos φ ) P (sin φ )
m
d (sin φ ) n
m
with k = 1 for m = 0, and k = 2 for m ≠ 0.
Zonal Harmonics
Top
2, 0 3, 0 4, 0
Side
5, 0
Sectoral Harmonics
Top
l=2 l=3 l=4
Side
l=5
Tesseral Harmonics
Top
2, 1 3, 1 3, 2 4, 1
Side
6, 4
Nodal Regression
Nodal Regression
Apsidal Rotation
Apsidal Rotation
Gravitational Effects
• Long ago when computers were slow…
• Gravitational modeling
– Often square gravity field truncations
• Appears the zonals contribute more
– Point to take away:

• Use “complete” field
• Any truncations should include additional, if not all,
zonal harmonics
Gravitational Modeling
• Satellite JERS (21867)
– Comparison to 12x12 field
– Note the variability over time
• 22x22 vs 18x18 and 70x22 vs 70x18
400.0 400.0
350.0 350.0
300.0 300.0
22x22 70x22
Difference (m)
Difference (m)
250.0 250.0
20x20 70x20
200.0 200.0
18x18 70x18
150.0 150.0
16x16 70x16
100.0 100.0
14x14 70x14
50.0 50.0
0.0 0.0
0.0 1440.0 2880.0 4320.0 5760.0 0.0 1440.0 2880.0 4320.0 5760.0
Time, min from Epoch Time, min from Epoch
Atmospheric Drag
• Large force for near-Earth satellites
– Very difficult to model
• Non-conservative force
– Total kinetic and potential energy not constant
• Heat, other losses through friction
K
K 1 c A 2 vrel
adrag = − ρ D vrel K
2 m vrel
Drag Effect on Orbits
Orbit tends to circularize
Available Data
300.0
F 10.7
ctr F 10.7
250.0
Trend
200.0
150.0
100.0
50.0
avg a p Solar Cycle Solar Cycle Solar Cycle Solar Cycle Solar Cycle
19 20 21 22 23
0.0
Jan-50 Jan-54 Jan-58 Jan-62 Jan-66 Jan-70 Jan-74 Jan-78 Jan-82 Jan-86 Jan-90 Jan-94 Jan-98 Jan-02 Jan-06 Jan-10
Space Weather – Predictions
• Lots of Variability
– Constant F10.7
• Not very accurate
• Never use 0.0!
– Schatten
• Varies with each solar cycle
– Polynomial Trend
– Matches several solar cycles
• F10.7 = 145 + 75*COS{ 0.001696 t + 0.35*SIN(0.001696 t )}
– t is the number of days from Jan 1, 1981
Observed vs Adjusted Solar Flux
• Data errors
40.0
– Some DRAO (obs) -

DRAO (adj)
data
30.0
inconsistencies
• 10-40 SFU 20.0
– Which does 10.0
the model 0.0
require? -10.0
• MSIS
-20.0
– Observed DRAO (adj) - DRAO (obs) -
Lenhart (adj)
Lenhart (adj)
data data
• Others -30.0
– Adjusted
-40.0
Jan-50 Jan-54 Jan-58 Jan-62 Jan-66 Jan-70 Jan-74 Jan-78 Jan-82 Jan-86 Jan-90 Jan-94 Jan-98 Jan-02 Jan-06
Solar Flux Predictions – Long Term
• Data differences
– One solar cycle 250
Schatten Oct 96
Mon Avg
• ~150 SFU Trend

Schatten Sep 97
Last F10.7
– Almost equal to the 200
Schatten Sep 00
solar min-max Schatten Jul 02
difference! 150
Schatten Jul 03
Schatten Mar 08
100
Schatten Nov 05
50
27-Nov-93 20-May-99 09-Nov-04 02-May-10 23-Oct-15 14-Apr-21 05-Oct-26
Now
Solar Flux Predictions – Shorter Term
• Data 300.0
differences Trend
250.0
– Min, Mid, and

Jan 94
Max 200.0
Jul 04
• 30-50 SFU 150.0
– Note timing of 100.0
Cycle is off
50.0
Solar Cycle 23, May 1996 - April 2000 - March 2008

0.0
Jan-94 Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12
Solar Flux Predictions – Shorter Term
• Early, Mid, and

Late 300.0
– Also 30-50 SFU

250.0
Trend
Apr 95
differences Mid
Apr 95
Late
Apr 95
200.0
Early
150.0
100.0
50.0
Solar Cycle 23, May 1996 - April 2000 - March 2008

0.0
Jan-94 Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10
Solar Flux Predictions – Short Term
• NOAA Predictions
– 27-day and 45-day (F10.7 and ap)
– 3-day
• 3-hourly Kp values off significantly as well
40
20
30
20 15
27-day F 10.7 27-day a p

10
Solar Flux Difference (SFU)
10
Geomagnetic a p Difference
0
45-day a p
-10
5
-20 45-day F 10.7
0
-30
-40
-5
-50
-60 -10
0 7 14 21 28 35 42 49 0 7 14 21 28 35 42 49
Prediction time (days) Prediction time (days)
Simulated Sensitivity Analysis
• JERS sample orbit
100000.0
– Different
J70
atmospheric models 10000.0 NRLMSIS-00
• Baseline J60
– Numerical 1000.0
MSIS86 MSIS90
propagation
Difference (m)
– Jacchia-Roberts 100.0
– 3-hourly J71
geomagnetic 10.0
– Relative comparison
only 1.0
0.1
0 1440 2880 4320 5760
Time, min from Epoch
Simulated Sensitivity Analysis
• JERS sample orbit
100000.0
– Different treatment ObsC81Dly

ObsConAll ObsDly
of the data 10000.0
ObsConAllAvg
Adj3Hr
– Baseline 1000.0
• Numerical
Propagation Difference (m)
100.0
ObsDly1700
• Jacchia-Roberts
10.0
• 3-hourly
Obs3HrSpl
geomagnetic 1.0
Obs3HrInt
0.1
0 1440 2880 4320 5760
NRLMSISE-00 Results – Short Term
1000.0
LOD20
CAS20 LAI20
CAT17 LOS20
CAT20 COS20
ObsConAllAvg
100.0 LAD20
CAD20
LOT20 LOS20
Difference (m)
LOI20
CAI20
L81ObsConAll COD20
10.0 COT17
COT20
LAS20
COI20
1.0
1320 1560 1800 2040 2280 2520 2760 3000 3240 3480
Time, min from Feb 20, 2008 00:00:00.000 UTC
NRLMSISE-00 Results – Long Term
100000.0
• Observations:
COT17
– Model specifies COT20
CAS20
observed 10000.0
LAD20 LAS20
• Adjusted ObsConAllAvg
ObsConAll
performed well LOS20
LAI20
1000.0
– Centered 81-day
Difference (m)
LOD20
LOT17
best LOT20
– 20:00 UTC best 100.0

CAD20
– Spline
interpolation very
good 10.0
COS20
• No single best CAT20

CAT17
LAS20 COD20
answer 1.0
CAI20
0 1440 2880 4320 5760 7200 8640 10080 11520

Jacchia-Roberts Results – Short Term
1000.0
LAT17 L81ObsConAll
LAS20
LAT20 LAD20
LAI20
LOS20 LOD20
100.0 LOT17 LOI20
LOT20
Difference (m)
COT17 CAT17
COS20
COD20 COT20
CAS20
COI20
CAT20
10.0
ObsConAllAvg
CAI20 CAD20
1.0
1320 1560 1800 2040 2280 2520 2760 3000 3240 3480
Jacchia-Roberts Results – Long Term
100000.0
• Observations:
– Adjusted
performed well in 10000.0
COT20
all cases COD20
LOS20
LAI20
– Centered 81-day
ObsConAll
best 1000.0
Difference (m)
ObsConAllAvg
– 20:00 UTC best
– Daily CAI20
CAD20
100.0
geomagnetic very
good, but all were
close
10.0
• No single best
answer
1.0
0 1440 2880 4320 5760 7200 8640 10080 11520
Atmospheric Drag Effects
• Atmospheric Drag
– Large variations
• Changing the atmospheric model
• Changing how the input data is interpreted
– F10.7 at 2000 UTC
– Last 81-day average F10.7 vs. the central 81-day average
– Using step functions for the atmospheric parameters vs
interpolation
– Many others (see AIAA and UC paper)
• 1-1000 km differences are possible
• Unable to determine if from data interpretation or model
differences
Third Body Forces
• Can affect GEO satellites strongly

• Conservative force (like gravity)
K K K
K G(m⊕ + msat )r⊕sat rsat3 r⊕3
a3−body = − + Gm3 ( − )
r⊕3 sat 3
rsat 3 r⊕33
Solar Radiation Pressure
• Large effect for high altitude satellites (GPS, GEO, etc)
– Non conservative force
• Shadowing by the Earth becomes very important
– All satellite altitudes
• Solar Irradiance (psr) is difficult to measure accurately
K
K cR ASun rsat−Sun
asrp = − pSR K
m rsat−Sun
Solar Irradiance (W/m2)
1376
1374
1372
1370
SMM
NOAA-9
NOAA-10
1368
Nimbus
Composite
1366 URS2
1364
1362
1360
1/1/1978 1/1/1980 1/1/1982 1/1/1984 1/1/1986 1/1/1988 1/1/1990 1/1/1992 1/1/1994 1/1/1996 1/1/1998 1/1/2000 1/1/2002
Earth Shadow Geometry
Earth Shadow Geometry
Solar Radiation Pressure Sensitivity
Results
• Solar Radiation Pressure
– Several variations shown 100.000
• 26690 (GPS) cylindrical
– Notice max is ~100m 10.000
– Definitions none
• Cylindrical
1.000 80.000
– Defines shadow type
Difference (m)
• App to true app to true
– Acct for light travel from 0.100
Sun to CB
• True 0.010
true
– Inst light from Sun

no
• No Boundary boundary
0.001
– Change step size at 0 1440 2880 4320 5760
penumbra/umbra Time, min from Epoch
– Point to take away

• Relatively small effect
• Some variations
Special Perturbations
• Numerically integrate the equations of motion
– Time consuming, but accurate
K μr K K K K K K K
a = 3 r + anon−spherical + adrag + a3−body + asrp + atides + aother
r
General Perturbations
• Truncate analytical expansions and solve directly
– Large time steps
• Each approach is mathematically different
– SGP4
– J2 only
– Other
Semianalytical
• Blend numerical and analytical
– Analytically solve secular and long period
components
– Numerically integrate the small short period
variations
• Fast and accurate
Force Model Sensitivity
Results
• Force model contributions
– Determine which forces contribute the largest effects
• 12x12 gravity field is the baseline
– Note
• Gravity and Drag are largest contributors for LEO
• 3rd body ~km effect for higher altitudes

• Trying to get the last cm from solid earth tides no good unless all
other forces are at least that precise
Force Model Contributions
• Low Earth Orbit

ISS JERS
25544 21867
1000000.0 1000000.0
vs Two- vs Two-
Body Body
100000.0 100000.0
vs EGM- vs EGM-
96 70x70 96 70x70
10000.0 10000.0
vs Drag vs Drag
MSIS 00 MSIS 00
Difference (m)
Difference (m)
1000.0 vs Drag 1000.0 vs Drag
Jrob Jrob
vs Third vs Third
100.0 Body 100.0 Body
vs SRP vs SRP
10.0 10.0
vs Solid vs Solid
Tides Tides
1.0 1.0
vs vs
Ocean Ocean
Tides Tides
0.1 0.1
0 1440 2880 4320 5760 0 1440 2880 4320 5760
• Low Earth Orbit

Starlette Vanguard II
7646 11
1000000.0 1000000.0
vs Two- vs Two-
Body 100000.0 Body
100000.0
vs EGM- vs EGM-
96 70x70 96 70x70
10000.0
10000.0
vs Drag vs Drag
MSIS 00 MSIS 00
Difference (m)
vs Drag 1000.0 vs Drag
Difference (m)
1000.0
Jrob Jrob
vs Third vs Third
100.0
100.0 Body Body
vs SRP vs SRP
10.0
10.0
vs Solid vs Solid
Tides Tides
1.0
1.0 vs
vs
Ocean Ocean
Tides Tides
0.1
0.1
0 1440 2880 4320 5760
0 1440 2880 4320 5760
Tim e, m in from Epoch Time, min from Epoch
• Low to Mid Earth Orbit

Topex GPS
22076 26690
1000000.0
1000000.0
vs Two- vs Two-
100000.0
Body 100000.0 Body
vs EGM- vs EGM-
96 70x70 96 70x70
10000.0
10000.0
vs Drag vs Drag
MSIS 00 MSIS 00
Difference (m)
Difference (m)
1000.0 1000.0
vs Drag vs Drag
Jrob Jrob
vs Third vs Third
100.0 100.0
Body Body
vs SRP vs SRP
10.0 10.0
vs Solid vs Solid
Tides Tides
1.0
1.0
vs vs
Ocean Ocean
Tides Tides
0.1
0.1
0 1440 2880 4320 5760
0 1440 2880 4320 5760
• Mid Earth Orbit, eccentric

SL 12 RB Molnyia
25054 20052
1000000.0
1000000.0
vs Two- vs Two-
100000.0 Body Body
100000.0
vs EGM- vs EGM-
96 70x70 96 70x70
10000.0
MSIS 00 MSIS 00
Difference (m)
1000.0 vs Drag
Difference (m)
vs Drag
1000.0
Jrob Jrob
vs Third vs Third
100.0
Body 100.0 Body
vs SRP vs SRP
10.0
10.0
vs Solid vs Solid
Tides Tides
1.0
vs 1.0 vs
Ocean Ocean
Tides Tides
0.1
0 1440 2880 4320 5760 0.1
0 1440 2880 4320 5760
• Low Earth and Geosynchronous Orbit

ISS (for comparison) Galaxy 11
25544 26038
1000000.0 1000000.0
vs Two- vs Two-
Body 100000.0 Body
100000.0
vs EGM- vs EGM-
96 70x70 96 70x70
10000.0 10000.0
vs Drag vs Drag
MSIS 00 MSIS 00
Difference (m)
Difference (m)

1000.0
Jrob Jrob
vs Third vs Third
100.0
100.0 Body Body
vs SRP vs SRP
10.0
10.0
vs Solid vs Solid
Tides Tides
1.0
1.0
vs vs
Ocean Ocean
Tides Tides
0.1
0.1
0 1440 2880 4320 5760
0 1440 2880 4320 5760
• Newton
• Kepler
• Perturbations
Chapter 10 • Orbit Determination
• Applications
Terms
– Process of determining an orbit from observations
– Also called Estimation
• Filtering
– Determining the current state after each observation
• Smoothing
– Improve previous state solutions using future data
– Runs backwards
Terms
• Deterministic
– Dynamics are known and can be calculated
– Propagation
• Assuming a specific set of force models
• Stochastic
– Uses observations to correct for unknown or
mis-modeled dynamics
Terms
• Least Squares
– Minimizes the sum-square of the residuals
– Depends on a fit span
• Length of time to process a batch of observations
– Often called Batch Least Squares (BLS)
Linear Least Squares Example
• Assume a mathematical model of motion

y = α + βx
• Residuals defined as
ri = yoi – yci = yoi – (α + β xoi)
• Cost function (Jacobian)
N N
J = ∑ ri 2 = f (α, β ) = ∑ ( yoi − (α + β xoi )) 2
i =1 i =1
• Minimization of residuals
Linear Least Squares Example
• Matrix development
⎡ 1 xo ⎤ ⎡ yo ⎤
⎢ 1⎥ ⎢ 1⎥
⎡1 1 ... 1 ⎤ ⎢ 1 xo1 ⎥ ⎡α ⎤ ⎡ 1 1 ... 1 ⎤ ⎢ yo2 ⎥
⎢ ⎥⎢ ⎥⎢ ⎥= ⎢ ⎥⎢ ⎥
⎢ xo
⎣ 1 xo2 ... xoN ⎥⎦ ⎢⎢... ... ⎥⎥ ⎢⎣ β ⎥⎦ ⎢⎣ xo1 xo2 ... xoN ⎥⎦ ⎢⎢ # ⎥⎥
⎢1 x ⎥ ⎢y ⎥
⎢⎣ o1 ⎥⎦ ⎢⎣ oN ⎥⎦
AT A X AT b
• Normal Equation
– X = (ATA)-1ATb
Drift ½ life
.. ... . . ........ .
. . ... .. ..
.... ..... ...... ... .. . . . . . ..... ...
.. ... .... .. .... . .... .. . . . . . . . . . .......... . . . ... . .. .
Noise . . ............. ............ .... . .. . . ... .. .. ..... ...... ..... .. ............................ ..
. . ..... .. .. . . .. .... . . . . ... . . . . . . ... .... . . .
Mean . . .. . .. .. . . .
. . ...... . .. ... . ... ........ ... .. .. ... ... .......... ...... .. . ...
Drift
. . . . Noise
. . .
. ... ... ........... .. . ...... .. ... .. .. .
.
...... . .... . .... ...... .. . . . .
.. ..... .... . . .. .
. . .. .
.
Bias . .
True Position

Statistical Concepts
• Dimensions and probability
Dimension z = 1j 2j 3j 4j
1 68.27 95.45 99.73 99.99 z

erf ( )
2
2 39.35 86.47 98.89 99.96 −z 2
1 − exp( )
2
3 19.87 73.85 97.07 99.89 z 2 −z 2

erf ( ) − z exp( )
2 π 2

Covariance Matrix
• Measure of uncertainty
• Grows with the satellite state propagation
P = (ATWA)-1
– W is weighting or sensor accuracies
– A is partial derivative matrix
• Correlation Coefficients
– Off diagonal terms
• Eigenvalues
– Indicates each axis of the ellipsoid

LS Applied to Satellites: Overview
Initial Orbit Obtain Good
Determination Initial State
Estimate, X
How good?
Radius of Curvature
What state representation?
Equinoctial, Keplerian, other
Propagate X to
observation
loop through times
Orbit observations
Determination
How to solve for Jacobian? Form Residuals

Analytical, finite differencing
Least Squares Solution method?
Classical, Single Value Decomposition
Solve
Jacobian
Converged? Solve Least

Squares

LS Algorithm: Matrix Inverse Approach
• FOR i = 1 to the number of observations (N)

– Propagate (SGP4, HPOP) nominal state to time of observation (TEME, ICRF)
– Find the slant range vector, sensor to the propagated state in the topocentric (SEZ)
coordinate system
– Determine nominal observations from the SEZ vector
– Find the b matrix as observed – nominal observations
– Form the A matrix
• Finite (or central) differences
• Analytical partials
– H, Partials depending on observation type
– Ф, Partials for state transition matrix.
– Accumulate ATWA and ATWb
• END FOR
• Find P = (ATWA)-1 using Gauss-Jordan elimination (LU decomposition and
back-substitution)
• Solve δx = P ATWb
• Check RMS for convergence
• Update state X = X + δx
• Repeat if not converged using updated state

Sequential Batch Least Squares
• Process additional observations

– Use previous results
• Bayes Theorem
• Normal Equation
– This is for “k” previously determined obs
– “k + n” new obs
T
δ x (0 | k + n ) = ( AnewWnew Anew + Pˆk−1 )−1 ( Anew
T
Wnewbnew + AkTWk bk )
Pˆ = Pˆ (0 | k + n ) = ( AT W A + Pˆ −1 )−1
k +n new new new k

Extended Kalman Filter
^ _
P1 P2 ^ _
P3 P4
^ ^ ^ ^ ^
dx2 dx d x4 dx5 P
^ 3 ^ 5
dx1 P4
^ _
P2 _ P5
_ P3
^ P1 Actual Orbit
P0
_ ^ _ _
^ _ ^ ^ ^ ^ _
X2 X2 X3 X3 X4 X5
X0 X1 X1 X4 X5

Extended Kalman Filter
∂z
at each obs time H k +1 =
∂Xˆ k +1
Prediction
tk +1
X (tk +1 , tk ) = ∫ X tk dt + X tk
tk

∂Xˆ tk+1
F=
∂Xˆ tk +1
(t , t ) = F (t )Φ(t , t )
Φ Predicted State
k +1 k k +1 k
δ xk +1 = 0 Predicted State Error

Pk +1 = ΦPk ΦT + Q Predicted Error Covariance
Update
b = z − H
k +1 k +1 X k +1
K k +1 = Pk +1 H kT+1[ H k +1 Pk +1 H kT+1 + R]−1 Kalman Gain
δ xˆ = δ x + K b
k +1 k +1 k +1 k +1
State Error Estimate
Pˆk +1 = Pk +1 − K k +1 H k +1 Pk +1 Error Covariance Estimate
Xˆ k +1 = X k +1 + δ xˆk +1
State Estimate

Averaging and Fit Spans
• Obs are taken periodically
• Updates often occur at regular intervals
• Least Squares approaches “average” data collected for
a “batch” of time – the Fit Span
Epoch 1 Epoch 2 Epoch 3
Obs Time
Fit span, 3 days
Daily Updates 1 2 3
Mean Change Short-periodic plus long-periodic, and secular
c Secular
Mean Change
Long-periodic and
secular
t1 t2 t3 t4
Time

• Newton
• Kepler
• Perturbations
Chapter 11
• Applications

Applications
• How do we put all this together and accomplish our
original goal?
– Many analyses possible
• Prediction
– Satellite look angles (Our original question)
• Behind the scenes

– Time of observations
– Coordinate systems throughout
– Orbit determination of observations to obtain a state vector
– Propagation to form an ephemeris
– Calculations for Sun and Satellite to determine visibility
– ...
– And several other smaller details!

Satellite Orbital Characteristics

Predicting Satellite Look Angles

Rise Set Characteristics

Finding the Site Information (1)
• Approximate formulation
– Non-rigorous ECEF
– Don’t account for sidereal/solar time differences
K K
ρ~ ECI = [ ROT3(−θ LST )][ROT2(−90° − φgd )]ρSEZ
K K
ρ ~ ECI = [ ROT3(−θ LST )][ROT2(−90° − φgd )]ρ SEZ
K K K
r~ ECI = ρ~ ECI + rSiteECI
K K K K
v~ ECI = ρ ~ ECI + ω⊕ × r~ ECI

Finding the Site Information (2)
• Rigorous formulation (STK approach)

– Precise ECEF
– Account for sidereal/solar time differences
K K
ρ ECEF = [ ROT 3(−λ )][ ROT 2(−90° − φ gd )]ρ SEZ
K K
ρ ECEF = [ ROT 3(−λ )][ ROT 2(−90° − φ gd )]ρ SEZ
K K K
rECEF = ρ ECEF + rSiteECEF
K K
vECEF = ρ ECEF
( yr , mon, day,UTC , ΔUT 1, ΔAT ) ⇒ (UT 1, TAI , TT , TUT 1 , TTT )
[ PREC ] = ROT 3(− z ) ROT 2(Θ) ROT 3(−ζ )
[ NUT ] = ROT 1(−ε ) ROT 3(− ΔΨ ) ROT 1(ε )
[ ST ] = ROT 3(θ AST )
[ PM ] = ROT 2(− x p ) ROT 1(− y p )
K K
Differences rECI = [ PREC ]T [ NUT ]T [ ST ]T [ PM ]T rECEF
from
K
{ K K K
vECI = [ PREC ]T [ NUT ]T [ ST ]T [ PM ]T vECEF + ω⊕ × rPEF }
Approximate

Results
• Rigorous approach
– Position (ECI)
• -5505.504883 km
• 56.449170
• 3821.871726
• Simplified approach
– Position (~ECI)
• -5503.79562 km
• 62.28191
• 3824.24480
• Difference
– 6.52 km
• Perhaps this is acceptable?
Impact
• Applying textbook solutions to real-world
problems will give the wrong answers
– Assumptions add up
– Examples:
• Communicating with a satellite using Laser comm
– At orbital velocity, 2 sec is nearly 14 km
» Will your signal be able to locate and receive?
• Will you pass System Acceptance Testing?

A word of Caution …
• Fundamentals vs Applications
– Undergraduate vs Graduate
– Classroom vs Operational
– Attention to detail important
– Nomenclature is important

Resources
• Book
– Microcosm
• Pam is here!
• http://www.celestrak.com/software/vallado-sw.asp
– TLE data
– EOP and Space Weather Data
– Code
• SGP4
• Other
– Errata
• Not all updated but most are
– Solutions
• Not complete – ask ☺

Questions??

Tutorial ASTRODYNAMICS PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tutorial ASTRODYNAMICS PDF

Uploaded by

Copyright:

Available Formats

Fundamentals of Astrodynamics

• Introduce you to the various topics that the text

• What we’ll need to understand

• Earth’s orbit is not exactly stable

Precession of Equinox Earth’s orbit

Celestial Reference Frame

IAU 1976 (zA ,θA, ζA)

IAU 1982 (∆ε, ∆ψ , ε0 , δ∆ε, δ ∆ψ ) ∆ε, ∆ψ

Sidereal Rotation R3[θGAST · 1982]

Polar Motion R2[-xp] R1[-yp]

Terrestrial Reference Frame

Celestial Reference Frame

MHB2000 (zA ,θA, ζA) MHB2000 (χA, ωA, ψA, ε0) X, Y X, Y

MHB2000 (∆ε, ∆ψ , ε0) X, Y, s

Bias-Precession [BPN] Eq. (xxx) = f ( X, Y, s )

Sidereal Rotation R3[θGAST · 2000 ] R3[θERA ]

Polar Motion R1[-yp] R2[-xp] R1[-yp] R2[-xp] R3[s’ ]

Terrestrial Reference Frame

Celestial Reference Frame

MHB2000 (∆ε, ∆ψ , ε0) X, Y, s

Bias-Precession [BPN] Eq. (xxx) = f ( X, Y, s )

Sidereal Rotation R3[θGAST · 2006 ] R3[θERA ]

Polar Motion R1[-yp] R2[-xp] R1[-yp] R2[-xp] R3[s’ ]

Terrestrial Reference Frame

• Find future position

Third Body Gravity

Solar Radiation Pressure

Solid Earth Tides

100 1000 10,00 100,000

Mean Change Short-periodic plus long-periodic, and secular

• Largest single contributor to the motion

l=2 l=3 l=4

– Point to take away:

Orbit tends to circularize

– Some DRAO (obs) -

– Which does 10.0

the model 0.0

• ~150 SFU Trend

– Almost equal to the 200

solar min-max Schatten Jul 02

– Min, Mid, and

• 30-50 SFU 150.0

– Note timing of 100.0

Solar Cycle 23, May 1996 - April 2000 - March 2008

• Early, Mid, and

– Also 30-50 SFU

Solar Cycle 23, May 1996 - April 2000 - March 2008

27-day F 10.7 27-day a p

-20 45-day F 10.7

– Different treatment ObsC81Dly

– 20:00 UTC best 100.0

• No single best CAT20

0 1440 2880 4320 5760 7200 8640 10080 11520

• Can affect GEO satellites strongly

• 26690 (GPS) cylindrical

– Notice max is ~100m 10.000

– Inst light from Sun

penumbra/umbra Time, min from Epoch

– Point to take away

– Point to take away:

• Low Earth Orbit

• Low Earth Orbit

• Low to Mid Earth Orbit