3.

9 Satellite State Representation

The motion of the satellite in its natural orbital plane has been discussed earlier in this chapter. This
section describes how to determine the position of a satellite in three dimensional space. In fact, we need
six quantities (known as orbital elements) for determining the position of a satellite in space.

Let P be a unit vector along the perihelion line. Suppose that the unit vector Q is perpendicular to P,
having a true anomaly of v = 90. The position vector r can be expresses in term of these unit vector as
follows:

r = x P + y Q,

As we have discussed in Chapter 2, the Fig 3.9 yields that

r = 𝑟𝑟𝑟𝑟 P + 𝑟𝑟𝑟𝑟 Q,

r = a (cos E  e) P + a√1 − 𝑟2 sin E Q,

Differentiating the above equation with respect to t, to find the velocity vector

𝑟̇ = −a sin E 𝑟̇ P + a√1 − 𝑟2 cos E 𝑟̇ Q, (3.54)

Eqs. (3.48) and (3.50) give that

√𝑟𝑟𝑟
𝑟𝑟̇ =
𝑟

Therefore, the velocity vector becomes

Let us now introduce the geocentric equatorial coordinate system for describing Earth-bound system. It is
assumed that Earth is positioned at the origin of the geocentric equatorial coordinate system with Z-axis
represent the axis of rotation of Earth and equatorial plane represents XY-reference plane. We
associated x-axis with vernal equinox  .
 Fig. 3.10: Celestial Sphere

We have the following relationship between the spherical coordinates: right ascension, declination 
and either the Cartesian coordinates (X, Y, Z).

𝑟 = [𝑟 𝑟 𝑟 ] = 𝑟 [𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 ]

with the following inverse transformation

𝑟 = √𝑟2 + 𝑟2 + 𝑟 2

𝑟
= 𝑟𝑟𝑟𝑟−1
𝑟

𝑟
= 𝑟𝑟𝑟𝑟−1
√𝑟2 +𝑟2

It is important to note for selecting appropriate quadrant that that the sign of the denominator (X) must be
same to the sign of cos . Following three angles are used for describing the orientation of orbital plane as
described by Montenbruck,2005.

(i) Inclination i (which measures the angle of intersection between the orbital plane and the
equator)
(ii) Right ascension of ascending node  (the angle between the vernal equinox and the
point on the orbit at which the satellite crosses the equator from south to north) and
 (iii) The argument of perigee  (angle between the direction of the ascending node and
the direction of the perigee).
We can express the position of a satellite in space as a function of these angles by sequence of three
Euler’s angle rotations. The unit vectors P, Q and W =h/h are used for defining the coordinates of the
orbital plane as described by Montenbruck,2005. Since,

(x, y, z) = (𝑟𝑟𝑟𝑟 , 𝑟𝑟𝑟𝑟 ,0)

Fig. 3.1: Equatorial plane configuration foe celestial objects.

If xyz coordinate system is rotated by an angle of  (i.e. with an x’-axis is directed to the ascending
node) around W, the coordinates are

(x, y, z) = (𝑟𝑟𝑟𝑟 (+) , 𝑟𝑟𝑟𝑟 (+) ,0)

and the corresponding transformation matrix is written as

𝑟[𝑟𝑟𝑟 (+) 𝑟𝑟𝑟 (+) 0 ] = 𝑟𝑟 (−) 𝑟[𝑟𝑟𝑟 𝑟𝑟𝑟 0 ]

Next, the x’-axis is rotated by an angle i to get equatorial coordinates counted from the line of nodes.
Finally the new z-axis is rotated by  then to obtain the equatorial coordinates counted from the
direction of equinox :
 [𝑟 𝑟 𝑟 ] = 𝑟𝑟 (−) 𝑟𝑟 (−𝑟) 𝑟𝑟 (−) 𝑟[𝑟𝑟𝑟 𝑟𝑟𝑟 0 ]

[𝑟 𝑟 𝑟 ] = 𝑟𝑟 (−) 𝑟𝑟 (−𝑟) 𝑟𝑟 (−) 𝑟[𝑟𝑟𝑟 𝑟𝑟𝑟 0 ]

[𝑟 𝑟 𝑟 ] = [𝑟𝑟𝑟 − 𝑟𝑟𝑟 0 𝑟𝑟𝑟 𝑟𝑟𝑟 0 0 0 1 ] [1 0 1 0 𝑟𝑟𝑟 𝑟 −

[𝑟 𝑟 𝑟 ] = 𝑟[𝑟𝑟𝑟 − 𝑟𝑟𝑟 0 𝑟𝑟𝑟 𝑟𝑟𝑟 0 0 0 1 ][1 0 0 0 𝑟𝑟𝑟 𝑟 −

[𝑟 𝑟 𝑟 ] = 𝑟[𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 +

with u = + v . The angle u is called argument of latitude which is the angle between the position line r and
the line of nodes. In particular, the vector P and Q are determined with a true anomaly of 0 and 90
respectively:

𝑟 = [𝑟𝑟𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 + 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 ]

𝑟 = [−𝑟𝑟𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟𝑟𝑟 + 𝑟𝑟𝑟𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 ]

We can compute the third vector W by taking cross product of P and Q as follows

𝑟 = [𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 ]

3.10 Orbit Elements from Position and Velocity

Six orbital elements (a, e, ʍ, ,,i) can determined if position r and velocity v of a satellite are known at
any instance. Since the angular momentum h is defined as follows:

h= r𝑟

h = [𝑟𝑟̇ − 𝑟𝑟̇ 𝑟𝑟̇ − 𝑟𝑟 𝑟𝑟̇ − 𝑟𝑟̇ ]

Since W =h/h and 𝑟 = [𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 ]

[𝑟𝑟 /𝑟 𝑟𝑟 /𝑟 𝑟𝑟 /𝑟 ] = [𝑟𝑟 𝑟𝑟 𝑟𝑟 ] = [𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 − 𝑟𝑟𝑟 𝑟 𝑟𝑟𝑟 𝑟𝑟𝑟 𝑟 ]

The inclination and right ascension of the ascending node can be determined as follows:

√𝑟2𝑟 +𝑟2𝑟
𝑟 = 𝑟𝑟𝑟𝑟−1
𝑟𝑟

−𝑟𝑟
= 𝑟𝑟𝑟𝑟−1
𝑟𝑟

The semi-latus rectum can be computed as

𝑟2
p= ,
𝑟𝑟

The semi-major axis can be expressed as

2 𝑟 2 −1
a=( − )
𝑟 𝑟𝑟

and the corresponding mean motion can determined using Eq. (3.51)

𝑟𝑟
n=√
𝑟3

The Eq. (3.32) for elliptic orbit yields

𝑟
e = √1 −
𝑟

The dot product r. 𝑟̇ can be computed using following two equations:

r = a (cos E  e) P + a√1 − 𝑟2 sin E Q, (3.54)

and

𝑟̇ = −a sin E 𝑟̇ P + a√1 − 𝑟2 cos E 𝑟̇ Q,

to prove that

r.𝑟̇ = 𝑟 2𝑟𝑟 𝑟𝑟𝑟 𝑟

It gives that

(𝑟.𝑟̇ )/𝑟2𝑟
𝑟 = 𝑟𝑟𝑟𝑟−1 𝑟
1−
𝑟
The eccentric anomaly can be obtained from Kepler’s Eq. (3.48) implies that

ʍ = E(t)  e sin E(t), (Montenbruck,2005).