Astro Dynamics Summary
Astro Dynamics Summary
Astro Dynamics Summary
Chapter 1
Fundamentals
1
1.4 Maneuvers with rocket thrust
If the maneuver is assumed to be an impulsive shot, thus an instant increase in velocity, the change in
velocity can be written as:
V 1 = V 0 + ∆V (1.3)
The change in angular momentum and change in orbital energy (per unit of mass) can be written as:
∆H = r0 × V 1 − r0 × V 0 = r0 × ∆V (1.4)
1 1 2
∆E = V12 − V02 = (∆V ) + V 0 • ∆V (1.5)
2 2
From the above equations some interesting conclusions can be drawn:
• For a given magnitude of ∆V , the maximum change in orbital angular momentum is achieved if the
impulsive shot is executed when the spacecraft is farthest away from Earth and if is perpendicular to
r0 .
• If the direction of the orbital angular momentum vector should not be changed, ∆V should be directed
in the initial orbital plane. If the direction of the angular momentum vector should be changed, a
component of ∆V should be directed perpendicular to the initial orbital plane.
• For a given magnitude of ∆V , the maximum change in (total) orbital energy is achieved if the impulsive
shot is executed at the point in the orbit where the velocity reaches a maximum value, and if ∆V is
directed along the velocity vector V 0 , i.e. tangentially to the (initial) orbit.
A finite burn time can be taken into account by adding a gravity loss and a drag loss, resulting in:
Z te
∆V = ∆Vid − ∆VG − ∆VD with ∆VG = g sin γdt (1.6)
t0
2. A line segment joining a planet and the Sun sweeps out equal areas during equal intervals of time.
3. The square of the orbital period of a planet is proportional to the cube of the semi-major axis of its
orbit.
2
AE4874 - Astrodynamics - Summary
Chapter 2
Let us consider a system composed of n bodies, which may be considered as point masses. When it is
assumed that outside the system of n bodies no other bodies exist, that no external forces act on the system,
and that within the system of n bodies only gravitational forces occur, then, applying Newton’s second law
of motion and Newton’s law of gravitation, the motion of body i with respect to the inertial reference frame
can be written as:
d2 ri X mi mj
mi 2
= G 3 rij (2.1)
dt rij
j6=i
The equation of motion of body i may be written as three scalar second-order differential equations.
Similarly, for the motion of n bodies 3n second-order differential equations can be written. Generally, this
set cannot be solved analytically and one has to rely on numerical integration techniques to determine the
motion of the bodies. However, some general characteristics of the many-body problem can be derived. These
characteristics are known as the ten integrals of motion, which will be derived in the following Section.
This shows that the center of mass is in constant motion. Integrating the differential equation results in the
first 6 integrals of motion, written in cartesian coordinates.
drcm
=a ; rcm = at + b (2.3)
dt
Three more integrals of motion can be found by taking the vector product of 2.1 and subsequently applying
a summation for all i.
X d2 ri X X mi mj
mi ri × 2 = G 3 ri × rj = 0 (2.4)
i
dt i
rij
j6=i
Resulting three integrals of motion, where H denotes the total angular momentum of the many-body system.
!
d X dri X dri
mi ri × =0 ; H= mi r i × =c (2.5)
dt i
dt i
dt
3
The vector H defines an invariable plane that passes through the center of mass of the n bodies and that is
perpendicular to the angular momentum vector. This plane is called the invariable plane of Laplace and
can be used as a reference plane for describing the motion of the n bodies.
The force field is non-central because the potential is dependent on the relative positions rij and not
the position with respect to an inertial reference frame ri . Secondly, the value of the potential at a fixed
position relative to the inertial reference frame will vary with time, because the bodies j are moving. For
such a time-varying potential the sum of kinetic and potential energy of body mi is not constant. Therefore,
we are dealing with a non-central, non-conservative force field.
The last integral of motion can be found by taking the scalar product of dri /dt and 2.1 and subsequently
applying a summation for all i. With Ek representing the total kinetic energy and Ep representing the total
potential energy.
X1 1 X X mi mj
mi Vi2 − G =C → εk + εp = C (2.6)
i
2 2 i
rij
j6=i
A stable system is a system for which d2 I/dt2 < 0, because even if the derivative is positive, eventually
the moment of inertia starts decreasing. So, a necessary, but not sufficient, condition for a stable system is:
C < 0. That this is a non-sufficient condition follows directly from the fact that, according to the above
equation, for given values of εk and εp , negative values of C can exist that still yield: d2 I/dt2 > 0.
In a stable systems no collisions and no escapes occur. In other words: all bodies stay within a finite distance
from the origin and the velocities of all bodies remain finite. In that case, the value of the expression between
brackets in will remain finite. Therefore, if the time interval is chosen large enough, the left-hand side of will
approach zero. So, for a sufficiently long averaging period, we find for a stable system:
1
4εk + 2εp = 0 → εk = − εp = −C (2.10)
2
4
AE4874 - Astrodynamics - Summary
Chapter 3
5
The following equations are used to derive the equations of motion for this representation.
m1
α= ; r13 = R + (1 − α)r12 ; r23 = R − αr12 ; R = αr13 + (1 − α)r23 (3.5)
m1 + m2
The following equations can be derived:
d2 R
r13 r23
= −GM α 3 + (1 − α) 3 (3.6)
dt2 r13 r23
d2 r12
r12 r13 r23
= −G (m1 + m 2 ) 3 + m 3 3 − 3 (3.7)
dt2 r12 r13 r23
form the Jacobi set of equations for the three-body problem. It is emphasized that these equations consti-
tute a twelfth-order system; the reduction from eighteenth order to twelfth order was essentially achieved by
the explicit use of the center-of-mass integrals. A further reduction is, of course, possible using the remaining
integrals of motion, the invariable plane of Laplace as reference plane and an angular coordinate to replace
time.
As an application of the Jacobi set of equations, we consider the so-called lunar case and planetary case.
In the lunar case, where P1 is the Earth, P2 the Moon and P3 the Sun, we know that:
d2 R d2 r12
R r12
= −GM α = −G(m1 + m2 ) 3 with α ≈ 1 ; r13 ≈ r23 ≈ R (3.8)
dt2 R3 dt2 r12
For the planetary case, with P1 the Sun, P2 the Earth and P3 a planet we arrive at the same approximative
equations of motion, using the following assumptions:
m3
α≈1 ; << 1 ; r13 ≈ R (3.9)
m1 + m2
An extinsive derivation beginning with the equations of motion leads to the following equations:
1 1 1 1 1 1
m2 r 1 × r 2 3 − r3 = 0 ; m3 r2 × r3 3 − r3 = 0 ; m1 r3 × r1 3 − r3 =0 (3.10)
r12 13 r23 12 r13 23
6
3.3 Circular restricted three-body problem
A partial solution of a special curcular restricted three-body problem is obtained by Lagrange. For this special
three-body problem, the following assumptions are made:
• The mass of two bodies is much larger than the mass of the third body. Then, the third body moves in
the gravity field of the two massive bodies, and the effect of the gravitational attraction by the third
body on the motion of these massive bodies can be neglected.
• The two massive bodies move in circular orbits about the center of mass of the system.
Figure 3.2: Inertial and rotating reference frames in the circular restricted three-body problem
The orbits of the two massive bodies (P1 , P2 ) being known, the problem is to determine the motion of the
third body P . The general three-body problem is thus reduced from nine second-order differential equations
to three second-order ones. This means a reduction from order eighteen to order six. Since the mass of the
third body is assumed to be negligible, the two main bodies move as if they form a two-body system.
A reference frame is chosen with its origin at the center of mass of the system of three bodies of which
the X-axis coincides with P1 P2 . This reference frame rotates with a constant angular velocity ω. Since both
massive bodies move in circular orbits about the center of mass O, we may conclude that the distances OP1
and OP2 are constant.
An extensive analysis can be performed resulting in the following equations:
∂U
ẍ − 2ẏ = (3.13)
∂x
∂U
ÿ + 2ẋ = (3.14)
∂y
∂U
z̈ = (3.15)
∂z
With:
1 2 1−µ µ
U= x + y2 + + (3.16)
2 r1 r2
From these equations we conclude that U is a potential function that accounts both for the gravitational
forces and for the centrifugal force. The potential function can, of course, not account for the Coriolis force,
because this force is a function of velocity components. The force field described by the potential U is clearly
non-central. Because the bodies P1 and P2 have fixed positions with respect to the rotating reference frame,
U is not explicitly a function of time, which means that the force field is conservative.
7
3.4 Jacobi’s integral
Using the potential function as described in the circular restricted problem an integral can be derived with
integration constant C. Where the value of C is determined by the position and velocity of body P .
V 2 = 2U − C (3.17)
2U = C (3.18)
This equation describes the surfaces of Hill. These are surfaces in XYZ-space on which the velocity of the
third body is zero.
Since for any real body V 2 > 0, the region in space where the third body can move is given by:
2U ≥ C (3.19)
So, although we cannot determine the orbit of the third body, we can determine which part of the XYZ-
space is accessible to the third body for a given value of C → initial conditions. In the figure below, this
unaccesible area is hatch for several values of C.
Figure 3.3: Schematic picture of the surfaces of HIll for creasing values of C
8
surfaces open up at certain points. These points are the Lagrange libration points. A minimum amount of
energy is needed to go from P1 to P2 through the Lagrange point L1.
Another way to describe the Lagrange libration points is to look at the potential function U which is
described in paragraph 3.3. The Lagrange libration points are the points where there is a local minimum in
the potential function U .
∂U ∂U ∂U
= = =0 (3.20)
∂x L ∂y L ∂z L
9
AE4874 - Astrodynamics - Summary
Chapter 4
An Equation of motion can be derived for the motion of body i under the influence of the gravitational
attraction of bodies j and a body k:
!
d2 rki m1 + m2 X rkj − rki rkj
= −G 3 rki + mj 3 − 3 (4.1)
dt2 rki rij rkj
j6=i,k
When the origin of the reference frame is body k, the equation becomes:
!
d2 ri m1 + m2 X rj − ri rj
= −G r i + mj − 3 (4.2)
dt2 ri3 3
rij rj
j6=i,k
10
For the magnitude of the perturbing acceleration follows from 4.1:
s
rid rd rid rd
ad = Gmd 3 − r3
rid
· 3 − r3
rid
(4.4)
d d
which gives: s
1 1 2rid rd cos β
ad = Gmd 4 + r4 −
rid rd3 rid
3 using A · B = |A| |B| cos θ (4.5)
d
11
AE4874 - Astrodynamics - Summary
Chapter 5
d2 r µ
2
= − 3r where µ = G(mk + mi ) (5.1)
dt r
dr d2 r
µ dr 1 d dr dr d µ 1 2 µ
· 2 + 3 ·r = · − =0 → V − =ε (5.2)
dt dt r dt 2 dt dt dt dt r 2 r
The area for a small surface element defined by the vectors r and r + ∆r is given in the equation below.
From this equation Kepler’s second law can be derived, which says that equal areas are swept out in equal
intervals of time.
1 2 dA 1 dϕ dA 1 dϕ H
∆A = r ∆ϕ + O(r∆r∆ϕ) → = r2 → = H ; = 2 (5.4)
2 dt 2 dt dt 2 dt r
d2 r
µ d dr dr dr µ
r · 2 + 3r · r = r· − · + =0 (5.5)
dt r dt dt dt dt r
12
The following derivatives are used:
∂r ∂ϕ H ∂r ∂ ṙ ∂ϕ H ∂ ṙ
ṙ = = 2 ; r̈ = = 2 (5.8)
∂ϕ ∂t r ∂ϕ ∂ϕ ∂t r ∂ϕ
The parameter u is introduced for simplicity, giving:
1 ∂r 1 ∂r ∂r ∂u 1 ∂u 1 ∂u ∂u
u= ; =− 2 ; = =− 2 → ṙ = Hu2 − 2 = −H (5.9)
r ∂u u ∂ϕ ∂u ∂ϕ u ∂ϕ u ∂ϕ ∂ϕ
∂2u µ
2
+u= 2 (5.11)
∂ϕ H
The solution to this differential equation is:
µ µ
u= + c1 cos ϕ + c2 sin ϕ = 2 (1 + c3 cos (ϕ + ω)) (5.12)
H2 H
Rewriting the solution gives the orbital equation:
H 2 /µ
r= (5.13)
1 + c3 cos (ϕ + ω)
Equation 5.13 is the general equation for a Conic section, which can be rewritten in the form given
below. With e as the eccentricity of the conic section and p as the semi-latus rectum. An example of a conic
section is an ellipse, therefore it can be concluded that this is the proof for Kepler’s first law.
p p
r= = (5.14)
1 + e cos (ϕ + ω) 1 + e cos θ
The type of conic section that is described by this equation depends on the eccentricity e, the following cases
can be distinguised:
e=0 : circle
0<e<1 : ellipse
e=1 : parabola
e>1 : hyperbola
Finally returning to the equation of motion. From this relation it can be concluded that the radial
acceleration of body i is equal to the difference between the centrifugal acceleration and the gravitational
acceleration.
µ
r̈ = rϕ̇2 − 2 (5.15)
r
13
5.3 Velocity components
When The flight path angle γ is introduced the velocity components can be rewritten. With ṙ as the radial
velocity and rθ̇ as the normal velocity.
Using the definition of the angular momentum H, the velocity components can be rewritten:
µ µ
ṙ = e sin θ ; rθ̇ = (1 + e cos θ) with H = rV cos γ (5.17)
H H
The flight path angle can be derived from the above equations, yielding:
ṙ e sin θ
tan γ = = (5.18)
rθ̇ 1 + e cos θ
The velocity components Vn and Vl can be derived using the following figure:
ṙ µe ṙ µ
Vl = = ; Vn = rθ̇ − = (5.19)
sin θ H tan θ H
These velocity components have a constant magnitude and the component Vl also has a constant orientation.
The component Vl is always oriented perpendicular to the axis of symmetry of the conic section. The
component Vn is always oriented in the direction of Vθ = rθ̇, which doesn’t have a constant orientation, but
is always perpedicular to the radius vector.
Using the figure and the velocity components Vn and Vl , the following equation can be derived:
µ 2 µe 2
ṙ2 + rθ̇ − = (5.20)
H H
This equation can be used to construct Velocity hodographs. These are displayed in the figure below:
14
Figure 5.2: Velocity components
15
A second phenomenon is related to the finite speed of light. This means that the sunlight intercepted by
body i at t0 is actually emitted by the Sun at t0 − ∆t. During the time interval ∆t light travels the distance
c∆t, while the body has moved over a distance rϕ̇∆t in the direction normal to the direction to the Sun
(Figure 5.12). This leads to the so-called aberration of the incoming sunlight; the aberration angle γ is given
by:
rϕ̇∆t rϕ̇
γ≈ = (5.27)
c∆t c
These equations show that the action of sunlight effectively reduces the central gravitational attraction force
by the Sun, but also produces two additional terms proportional to the radial and circumferential velocities
of the body. The second additional term corresponds to a drag-type of force. These adjusted equations of
motion were first formulated by H.P. Robertson.
16
AE4874 - Astrodynamics - Summary
Chapter 6
17
6.1 Circular velocity
The circular velocity is derived by using the definition of a circular orbit. Where the eccentricity e is equal to
zero and the semi major axis a is equal to the radius r. It can also be derived using the eccentricity vector.
The circular velocity is given by: r
µ
Vc = (6.5)
r
With the use of the definitions of an ellipse the following equation can be derived:
PG b p p
0
= = 1 − e2 ; r sin θ = a 1 − e2 sin E (6.7)
P G a
Using the equations above, the following equation can be derived by substituting:
Combining all the equations and using a trigonometric expression, the relationship between E and θ can be
derived: r
θ 1+e E θ 1 − cos θ
tan = tan ; tan2 = (6.9)
2 1−e 2 2 1 + cos θ
Differentiating Equation 6.8 and substituting the earlier found equations result in:
r
µ
E − e sin E = n(t − τ ) = M with n = (6.10)
a3
18
AE4874 - Astrodynamics - Summary
Chapter 7
Parabolic orbits
With these relations the general equation for a parabola can be derived:
p
r= (7.2)
1 + cos θ
A parabolic orbit escapes the gravitation of the attracting body. The escape velocity can thus be derived by
using e = 1 and the eccentricity vector. Or by using the energy equation and substituting the earlier given
conditions:
2µ √
r
Vesc = = 2Vc (7.3)
r
√
In other words the escape velocity can be computed by multiplying the local circular velocity with 2. It
can be derived that the total energy of a body in a parabolic orbit is always equal to zero.
ε=0 (7.4)
19
AE4874 - Astrodynamics - Summary
Chapter 8
Hyperbolic orbits
The same equations as for the elliptical orbit hold for the hyperbolic orbit. The following definitions hold
for the hyperbolic orbit:
e>1 ; a<0 (8.1)
With the orbit equation, the limit value of θ can be derived by setting the radius r to ∞.
a(1 − e2 )
r= ; θlim = arccos(−1/e) (8.2)
1 + e cos θ
Because for the hyperbolic orbit the eccentricity is negative, we find that the total energy is positive. This
can be seen as that the kinetic energy of the body is larger than the potential(graviational) energy at any
point in the orbit.
ε>0 (8.3)
The velocity reaches a minimum for r = ∞, giving the excess velocity.
r
µ
V∞ = − (8.4)
a
Using the definition of the escape velocity and the excess velocity the following equation can be derived:
V 2 = Vesc
2 2
+ V∞ (8.5)
20
AE4874 - Astrodynamics - Summary
Chapter 9
Relative motion
The motion of a satellite 2 w.r.t satellite 1 can be determined by adjusting the equations of motion. The
reference frame that is used is given in the figure below. Where the X,Y and Z specify the radial, along-track
and cross-track directions of the motion of satellite 2 with respect to satellite 1.
21
AE4874 - Astrodynamics - Summary
Chapter 10
A reference system then is the complete specification of how a celestial coordinate system is to be formed.
Both the origin and the orientation of the fundamental (reference) planes (or axes) are defined. A reference
system also incorporates a specification of the fundamental models needed to construct the system; that
is, the basis for the algorithms used to transform between observable quantities and reference data in the
system.
A reference frame, on the other hand, consists of a set of identifiable fiducial points on the sky along
with their coordinates, which serves as the practical realization of a reference system.
22
20/21 and September 22/23 each year. These crossing points are therefore called the vernal equinox
and the autumnal equinox, respectively.
• Declination: Geocentric angle, measured from the celestial equator, along the object’s hour circle, to
the object.
• Right ascension: Geocentric angle, measured along the celestial equator, from the vernal equinox to
the foot of the object’s hour circle.
10.3 Time
• Solar time: Time defined by the angular distance covered by the Sun on the celestial sphere after its
last crossing of the observer’s celestial meridian.
• Sidereal time: Time defined by the angular distance covered by the vernal equinox on the celestial
sphere after its last crossing of the observer’s celestial meridian.
• Atomic time (TAI): Time based on the analysis of about 200 frequency standards (atomic clocks)
maintained by several countries to keep a unit of time as close to the ideal SI second as possible
(defined in terms of Caesium 133 transitions).
• Universal time (UT / UT1): A standardised mean solar time, based on a fictitious mean Sun that
moves at a uniform rate eastward along the celestial equator.
• Universal time coordinated (UTC): A hybrid standard of time, of which the progression is determined
by atomic time (TAI), but leap seconds are introduced, when needed, to keep up with Universal Time
(UT1).
23
AE4874 - Astrodynamics - Summary
Chapter 11
Equations to remember
2µ √
r
Vesc = = 2Vc ; a=∞ ; e=1 (11.7)
a
Excess velocity (Hyperbolic orbit):
r
µ
V∞ = ; V 2 = V∞
2 2
+ Vesc ; r=∞ ; a<0 ; e>1 (11.8)
−a
Eccentric anomaly, E:
24