Skip to content

Classical Orbital Elements and Coordinate Transformations

This appendix provides the complete mathematical reference for classical orbital mechanics, coordinate frames, and the transformations used throughout the book. All derivations follow the conventions of Bate, Mueller, and White (1971) and Vallado (2013).

Standard Gravitational Parameters

The standard gravitational parameter \(\mu = GM\) combines the universal gravitational constant \(G = 6.674\,30 \times 10^{-11}\) m\(^3\) kg\(^{-1}\) s\(^{-2}\) with the mass \(M\) of the central body. Using \(\mu\) rather than \(G\) and \(M\) separately eliminates the relative uncertainty in \(G\) from precision orbit determination.

Table 1. Standard gravitational parameters for solar-system bodies
Body \(\mu\) (km\(^3\) s\(^{-2}\)) \(\mu\) (m\(^3\) s\(^{-2}\)) Source
Sun \(1.327\,124\,400\,18 \times 10^{11}\) \(1.327\,124\,400\,18 \times 10^{20}\) IAU 2012
Earth \(3.986\,004\,418 \times 10^{5}\) \(3.986\,004\,418 \times 10^{14}\) EGM2008
Moon \(4.902\,800\,066 \times 10^{3}\) \(4.902\,800\,066 \times 10^{12}\) DE430
Mars \(4.282\,837\,59 \times 10^{4}\) \(4.282\,837\,59 \times 10^{13}\) MRO/MGS
Jupiter \(1.266\,865\,349 \times 10^{8}\) \(1.266\,865\,349 \times 10^{17}\) Juno
Saturn \(3.793\,120\,62 \times 10^{7}\) \(3.793\,120\,62 \times 10^{16}\) Cassini
Venus \(3.248\,585\,92 \times 10^{5}\) \(3.248\,585\,92 \times 10^{14}\) MESSENGER
Mercury \(2.203\,186\,86 \times 10^{4}\) \(2.203\,186\,86 \times 10^{13}\) MESSENGER

Earth Physical Constants

Table 2 lists the primary physical constants of the Earth used in satellite orbit computation, attitude determination, and communications link analyses.

Table 2. Earth physical constants (WGS-84 / EGM2008 values unless noted)
Constant Symbol Value Units
Equatorial radius \(R_E\) 6 378.137 0 km
Polar radius \(R_p\) 6 356.752 3 km
Flattening \(f\) 1/298.257 223 563 –
Mean radius \(\bar{R}_E\) 6 371.008 8 km
Grav. parameter \(\mu_E\) 398 600.441 8 km\(^3\) s\(^{-2}\)
\(J_2\) zonal harmonic \(J_2\) \(1.082\,626\,68 \times 10^{-3}\) –
\(J_3\) zonal harmonic \(J_3\) \(-2.532\,435 \times 10^{-6}\) –
\(J_4\) zonal harmonic \(J_4\) \(-1.619\,331 \times 10^{-6}\) –
\(J_5\) zonal harmonic \(J_5\) \(-2.277\,161 \times 10^{-7}\) –
Earth rotation rate \(\omega_E\) \(7.292\,115\,0 \times 10^{-5}\) rad s\(^{-1}\)
Speed of light (vacuum) \(c\) 299 792 458 m s\(^{-1}\)
Standard gravity \(g_0\) 9.806 65 m s\(^{-2}\)
Solar pressure constant \(P_\odot\) \(4.560 \times 10^{-6}\) N m\(^{-2}\)
Sidereal day \(T_\text{sid}\) 86 164.090 5 s
Solar day \(T_\odot\) 86 400.000 0 s

The Six Classical Orbital Elements

Any two-body Keplerian orbit is uniquely determined by six scalar constants—the Classical Orbital Elements (COE), also called Keplerian elements. Three elements define the shape and size of the ellipse; three orient that ellipse in inertial space.

Size and Shape

\(a\) – Semi-major axis (km).
Half the longest diameter of the orbital ellipse. It sets the orbital energy \(\mathcal{E} = -\mu/(2a)\) and the period \(T = 2\pi\sqrt{a^3/\mu}\).
\(e\) – Eccentricity (dimensionless).
Measures the deviation from a circle. Ranges from \(0\) (circular) through \(<1\) (elliptical) to \(1\) (parabolic) and \(>1\) (hyperbolic). Related to the semi-latus rectum by \(p = a(1-e^2)\).
\(\nu\) – True anomaly (rad or deg).
Angle measured in the orbital plane from the perigee direction to the satellite, in the direction of motion. It is the position element; all five others are constants of integration.

Orientation in Inertial Space

\(i\) – Inclination (rad or deg).
Angle between the orbital plane and the equatorial plane, measured at the ascending node. \(0 \le i \le \pi\). Prograde orbits have \(i < 90\degree{}\); retrograde have \(i > 90\degree{}\).
\(\Omega\) – Right Ascension of the Ascending Node (RAAN) (rad or deg).
Angle in the equatorial plane from the vernal equinox direction (\(\hat{\boldsymbol{X}}\)) eastward to the ascending node. Ranges from 0 to \(2\pi\).
\(\omega\) – Argument of Perigee (AoP) (rad or deg).
Angle in the orbital plane from the ascending node to perigee, measured in the direction of motion. Undefined for circular orbits (\(e=0\)).
Classical orbital elements and their geometric relationship to the Earth-Centred Inertial (ECI) frame.
Figure 1. Classical orbital elements and their geometric relationship to the Earth-Centred Inertial (ECI) frame.

Anomaly Conversions

True Anomaly to Eccentric Anomaly

The eccentric anomaly \(E\) is an auxiliary angle defined on a circumscribed circle of radius \(a\). The conversion is:

\[\tan\!\left(\frac{E}{2}\right) = \sqrt{\frac{1-e}{1+e}}\,\tan\!\left(\frac{\nu}{2}\right)\tag{1}\]

Careful use of the atan2 function is required to resolve the quadrant ambiguity: \(E\) and \(\nu\) always lie in the same half-plane (\(0 \le \nu < \pi \Leftrightarrow 0 \le E < \pi\)).

Eccentric Anomaly to Mean Anomaly (Kepler’s Equation)

The mean anomaly \(M\) grows linearly in time:

\[M = E - e\sin E, \qquad M = n(t - t_p)\tag{2}\]

where \(n = \sqrt{\mu/a^3}\) is the mean motion and \(t_p\) is the time of perigee passage.

Mean Anomaly to Eccentric Anomaly: Newton–Raphson Solution

Kepler’s equation (2) cannot be inverted analytically; numerical methods are required. The Newton–Raphson iteration is:

\[\begin{align*} f(E) &= E - e\sin E - M = 0 \notag\\ f'(E) &= 1 - e\cos E \notag\\ E_{k+1} &= E_k - \frac{E_k - e\sin E_k - M}{1 - e\cos E_k} \end{align*}\]

Algorithm A.1 – Kepler Equation Solver

  1. Set \(E_0 = M\) for \(e < 0.8\), or \(E_0 = \pi\) for \(e \ge 0.8\).

  2. Iterate (3) until \(|E_{k+1} - E_k| < 10^{-12}\) rad (typically 3–6 iterations for \(e < 0.97\)).

  3. Return \(E\) modulo \(2\pi\).

For near-parabolic orbits (\(e > 0.999\)) the Battin (1987) universal variable formulation is recommended.

Position and Velocity in the Perifocal Frame

The perifocal (PQW) frame has unit vectors: \(\hat{\boldsymbol{p}}\) pointing toward perigee, \(\hat{\boldsymbol{q}}\) in the orbital plane \(90\degree{}\) ahead of perigee (in the direction of motion), and \(\hat{\boldsymbol{w}} = \hat{\boldsymbol{p}}\times\hat{\boldsymbol{q}}\) normal to the orbital plane.

\[\begin{align*} \boldsymbol{r}_{\text{peri}} &= \frac{p}{1 + e\cos\nu} \begin{pmatrix} \cos\nu \\ \sin\nu \\ 0 \end{pmatrix} \\[6pt] \boldsymbol{v}_{\text{peri}} &= \sqrt{\frac{\mu}{p}} \begin{pmatrix} -\sin\nu \\ e + \cos\nu \\ 0 \end{pmatrix} \end{align*}\]

where the semi-latus rectum \(p = a(1-e^2)\).

Rotation Matrices and ECI State Vector

Elementary Rotation Matrices

Positive rotations follow the right-hand rule about axis \(k\):

\[\begin{align*} \boldsymbol{R}_1(\theta) &= \begin{pmatrix} 1 & 0 & 0 \\ 0 & \cos\theta & \sin\theta \\ 0 & -\sin\theta & \cos\theta \end{pmatrix} \\[8pt] \boldsymbol{R}_2(\theta) &= \begin{pmatrix} \cos\theta & 0 & -\sin\theta \\ 0 & 1 & 0 \\ \sin\theta & 0 & \cos\theta \end{pmatrix} \\[8pt] \boldsymbol{R}_3(\theta) &= \begin{pmatrix} \cos\theta & \sin\theta & 0 \\ -\sin\theta & \cos\theta & 0 \\ 0 & 0 & 1 \end{pmatrix} \end{align*}\]

Perifocal to ECI Transformation

The ECI position vector is obtained by three successive rotations that undo the three orientation angles:

\[\boldsymbol{r}_{\text{ECI}} = \boldsymbol{R}_3(-\Omega)\,\boldsymbol{R}_1(-i)\,\boldsymbol{R}_3(-\omega)\,\boldsymbol{r}_{\text{peri}}\tag{9}\]

Writing out the composite direction cosine matrix \(\boldsymbol{Q} = \boldsymbol{R}_3(-\Omega)\,\boldsymbol{R}_1(-i)\,\boldsymbol{R}_3(-\omega)\):

\[Q_{ij} = \begin{pmatrix} \cos\Omega\cos\omega - \sin\Omega\sin\omega\cos i & -\cos\Omega\sin\omega - \sin\Omega\cos\omega\cos i & \sin\Omega\sin i \\ \sin\Omega\cos\omega + \cos\Omega\sin\omega\cos i & -\sin\Omega\sin\omega + \cos\Omega\cos\omega\cos i & -\cos\Omega\sin i \\ \sin\omega\sin i & \cos\omega\sin i & \cos i \end{pmatrix}\tag{10}\]

The same rotation applies to velocity: \(\boldsymbol{v}_{\text{ECI}} = \boldsymbol{Q}\,\boldsymbol{v}_{\text{peri}}\).

State Vector to Orbital Elements (Inverse Problem)

Given position \(\boldsymbol{r}_{\text{ECI}}\) and velocity \(\boldsymbol{v}_{\text{ECI}}\) in ECI coordinates, the classical elements follow from:

Algorithm A.2 – State Vector to Classical Orbital Elements

  1. Compute specific angular momentum: \(\boldsymbol{h} = \boldsymbol{r} \times \boldsymbol{v}\), \(h = |\boldsymbol{h}|\).

  2. Compute node vector: \(\boldsymbol{N} = \hat{\boldsymbol{K}} \times \boldsymbol{h}\), \(N = |\boldsymbol{N}|\).

  3. Compute eccentricity vector: \(\boldsymbol{e}_v = \dfrac{1}{\mu}\!\left[ \left(v^2 - \dfrac{\mu}{r}\right)\boldsymbol{r} - (\boldsymbol{r}\cdot\boldsymbol{v})\boldsymbol{v} \right]\), \(e = |\boldsymbol{e}_v|\).

  4. Compute specific mechanical energy: \(\mathcal{E} = v^2/2 - \mu/r\).

  5. Semi-major axis: \(a = -\mu/(2\mathcal{E})\).

  6. Inclination: \(i = \arccos(h_z/h)\).

  7. RAAN: \(\Omega = \arccos(N_x/N)\). If \(N_y < 0\), then \(\Omega = 2\pi - \Omega\).

  8. Argument of perigee: \(\omega = \arccos(\boldsymbol{N}\cdot\boldsymbol{e}_v/(Ne))\). If \(e_{v,z} < 0\), then \(\omega = 2\pi - \omega\).

  9. True anomaly: \(\nu = \arccos(\boldsymbol{e}_v\cdot\boldsymbol{r}/(er))\). If \(\boldsymbol{r}\cdot\boldsymbol{v} < 0\), then \(\nu = 2\pi - \nu\).

Special cases: circular orbits (\(e < 10^{-4}\)) use the argument of latitude \(u = \omega + \nu\); equatorial orbits (\(i < 0.001\degree{}\)) use the true longitude of perigee \(\tilde{\omega} = \Omega + \omega\).

ECI to ECEF Transformation

The Earth-Centred Earth-Fixed (ECEF) frame rotates with the Earth. The transformation from ECI to ECEF requires accounting for Earth rotation, precession, nutation, and polar motion. For satellite orbit prediction to \(<1\) km accuracy the following simplified form is sufficient:

\[\boldsymbol{r}_{\text{ECEF}} = \boldsymbol{R}_3(\theta_{\text{GAST}})\,\boldsymbol{r}_{\text{ECI}}\tag{11}\]

Greenwich Apparent Sidereal Time (GAST)

GAST at epoch \(t\) (Julian date \(J\)) is computed as follows. Define the Julian centuries from J2000.0:

\[T_u = \frac{J - 2\,451\,545.0}{36\,525}\tag{12}\]

Earth Rotation Angle (ERA):

\[\theta_{\text{ERA}} = 2\pi\left(0.779\,057\,273\,264\,0 + 1.002\,737\,811\,911\,354\,48\,D_u\right)\tag{13}\]

where \(D_u = J - 2\,451\,545.0\) is the number of UT1 days from J2000.0. Adding the equation of the equinoxes \(\Delta\psi\cos\varepsilon\) gives GAST. For most engineering applications, Greenwich Mean Sidereal Time (GMST) is adequate:

\[\theta_{\text{GMST}} = 280.461\,618\,375 + 360.985\,647\,366\,29\,D_u \quad\text{(degrees, modulo 360)}\tag{14}\]

ECEF to Geodetic Coordinates

Given ECEF position \((X, Y, Z)\), geodetic longitude \(\lambda\), latitude \(\phi\), and altitude \(h\) are found as follows.

\[\lambda = \mathrm{atan2}(Y, X)\tag{15}\]

Geodetic latitude and altitude require iteration because the Earth is an oblate spheroid. Define:

\[\begin{align*} e_E^2 &= 2f - f^2 = 6.694\,379\,990\,14 \times 10^{-3} \quad (\text{first eccentricity squared}) \\ p &= \sqrt{X^2 + Y^2} \end{align*}\]

Bowring’s Iterative Method:

  1. Initial estimate: \(\phi_0 = \mathrm{atan2}(Z, p(1-e_E^2))\).

  2. Compute radius of curvature in prime vertical: \(N = R_E / \sqrt{1 - e_E^2 \sin^2\phi}\).

  3. Update: \(\phi = \mathrm{atan2}\!\left(Z + e_E^2 N\sin\phi,\; p\right)\).

  4. Repeat steps 2–3 until \(|\Delta\phi| < 10^{-11}\) rad (3 iterations converge to sub-millimetre accuracy).

  5. Altitude: \(h = p/\cos\phi - N\). At poles (\(\phi = \pm 90\degree{}\)): \(h = |Z|/\sin\phi - N(1-e_E^2)\).

Two-Line Element Set (TLE) Format

The TLE format, standardised by USSTRATCOM/Space-Track, encodes a near-earth satellite state compatible with the SGP4/SDP4 propagators.

Table 3. TLE field definitions (Line 1)
Col. Field Example Description
1 Line number 1 Always 1
3–7 Satellite number 25544 NORAD catalogue ID
8 Classification U U=Unclassified, C=Classified
10–11 Int. desig. year 98 Launch year (2-digit)
12–14 Int. desig. launch 067 Launch number in year
15–17 Int. desig. piece A Piece of launch
19–20 Epoch year 24 Year (2-digit)
21–32 Epoch day 264.51782528 Day of year (fractional)
34–43 Ballistic coeff. .00001764 First deriv. of mean motion/2 (rev/day\(^2\))
45–52 2nd deriv. mean mo. 00000+0 Second deriv./6 (rev/day\(^3\))
54–61 BSTAR drag term 31088-3 SGP4 drag coefficient (1/R\(_E\))
63 Element set type 0 0 = distributed to public
65–68 Element set no. 999 Incrementing integer
69 Checksum 3 Modulo-10 checksum
Table 4. TLE field definitions (Line 2)
Col. Field Example Description
1 Line number 2 Always 2
3–7 Satellite number 25544 Same as Line 1
9–16 Inclination 051.6400 Degrees
18–25 RAAN 254.6611 Degrees
27–33 Eccentricity 0001764 Decimal point assumed at start
35–42 Arg. of perigee 148.9962 Degrees
44–51 Mean anomaly 211.2021 Degrees
53–63 Mean motion 15.49560679 Revolutions per day
64–68 Rev. at epoch 56353 Revolution number
69 Checksum 7 Modulo-10 checksum

TLE Epoch Convention

The TLE epoch day is the fractional day of year in UTC. To convert to a Julian date:

\[J_{\text{epoch}} = J_{\text{Jan 0.0}} + \text{Day}_{\text{frac}}\]

where \(J_{\text{Jan 0.0}}\) is the Julian date of 0 January of the epoch year (equivalently, 31 December of the previous year at 00:00:00 UTC).

Mean Motion and Semi-major Axis

The TLE mean motion \(n_0\) (rev/day) relates to the osculating semi-major axis via the SGP4 Kozai-mean element definition, not a simple Keplerian relation. The approximate Keplerian relation is:

\[a = \left(\frac{\mu}{n^2}\right)^{1/3}, \quad n = n_0 \cdot \frac{2\pi}{86\,400}\;\text{rad/s}\tag{19}\]

Checksum Calculation

Each TLE line ends with a modulo-10 checksum. Sum the integer value of each digit character; treat each minus sign as 1, spaces and letters as 0. Take the result modulo 10.

Common Orbit Types: Reference Elements

Table 5 lists nominal orbital elements for representative satellites. All values are approximate; use current TLEs for operational calculations.

Table 5. Nominal orbital elements for common satellite types (epoch \(\approx\) 2024)
Satellite / Type \(a\) \(e\) \(i\) \(\Omega\) \(\omega\) \(n\) Period
(km) (deg) (deg) (deg) (rev/d)
ISS (LEO) 6 797 0.0004 51.64 – – 15.49 93.0 min
Starlink (550 km shell) 6 921 0.0001 53.0 – – 15.08 95.5 min
Envisat (SSO) 7 159 0.0001 98.55 – – 14.33 100.5 min
GPS Block IIF 26 560 0.0001 55.0 – – 2.006 11.97 h
GLONASS-M 25 508 0.001 64.8 – – 2.131 11.26 h
Galileo FOC 29 600 0.0003 56.0 – – 1.705 14.08 h
Molniya (HEO) 26 556 0.72 63.4 – 270.0 2.006 11.97 h
GOES-16 (GEO) 42 164 \(<0.001\) \(<0.1\) – – 1.003 23.93 h
Meteosat-12 (GEO) 42 164 \(<0.001\) \(<0.1\) – – 1.003 23.93 h
\(a\) = semi-major axis from Earth’s centre; \(n\) = mean motion; – = varies by orbital plane.