LAMBERTS PROBLEM

Interplanetary Mission Design

Kate Davis
INTRODUCTION
Lamberts problem is a way to solve for the trajectory connecting two position vectors with a given time
of ight. In Figure 1, r
0
and r
f
dene the positions of the initial planet (P
0
) at the time of departure and
the nal planet (P
f
) at the time of arrival. With these positions and the time of ight, solving Lamberts
problem will dene the orbital elements of the transfer orbit. Once the orbital elements of the transfer orbit
are computed, the velocities at departure and arrival may be found.
Initial Planet, P0
at time t0
Final Planet, Pf
at time tf
r
0
r
f

t = t
f
t
0
Transfer Time: t
Figure 1. Lamberts Problem
Lamberts Theorem
According to Lamberts Theorem, the transfer time t from P
0
to P
f
is independent of the orbits eccen-
tricity and depends only on the sum of the magnitudes of the position vectors, the semimajor axis a and the
length of the chord joining P
0
and P
f
.
There are many solutions to Lamberts problem that may be found in literature. This handout will detail
the Universal Variables Formulation. The denitions of the variables used in this handout may be found in
Table 1.
1
Table 1. Notation
a = Semimajor axis
e = Eccentricity
E = Eccentric Anomaly
i = Inclination
k = Integer number of revolutions
r = Position Vector Magnitude (r = r/|r|)
t = Time of ight
= Gravitational Parameter
= True Anomaly
FORMULATION
Begin with the general form of Keplers equation:
t
f
t
0
= t =

a
3

_
2k + (E
f
e sin E
f
) (E
0
e sin E
0
)
_
, (1)
where the subscripts 0 and f refer to the initial and nal states, respectively. In this formulation, we
are not concerned with multiple revolutions, so we will ignore the 2k term. The equation is rearranged to
obtain:

t =

a
3
_
E + e (sin E
0
sin E
f
)

t =

a
3
E +

a
3
e (sin E
0
sin E
f
) . (2)
We will the use universal variable to rearrange Equation 2 into quantities that are already known.
Although not provided here, Bate, Mueller, & White
1
and Vallado
2
provide detailed explanations on how to
derive the universal variable . The expression for is
=

a (E
f
E
0
) =

a E. (3)
Furthermore, dene the constants c
2
and c
3
as
c
2
=
1 cos E
E
2
(4)
c
3
=
E sin E
E
3
. (5)
The quantity
3
c
3
is computed and rearranged:

3
c
3
=
_
aE
_
3
E sin E
E
3

3
c
3
=

a
3
E

a
3
sin E

a
3
E =
3
c
3
+

a
3
sin E. (6)
Insert Equation 6 into Equation 2 to obtain:

t =
3
c
3
+

a
3
sin E +

a
3
e (sin E
0
sin E
f
) . (7)
2
Use the trigonometric identity
sin E = sin E
f
cos E
0
cos E
f
sin E
0
, (8)
and Equation 7 becomes

t =
3
c
3
+

a
3
_
sin E
f
cos E
0
cos E
f
sin E
0
+ e sin E
0
e sin E
f
_

t =
3
c
3
+

a
3
_
sin E
0
(e cos E
f
) sin E
f
(e cos E
0
)

. (9)
The sine and cosine functions for true anomaly are given by:
cos =
e cos E
e cos E 1
(10)
sin =

1 e
2
sin E
1 e cos E
(11)
We wish to express the terms in Equation 9 in terms of the eccentric anomaly and the true anomaly. There-
fore, we will multiply the nal term in Equation 9 by where:
= 1 =

1 e
2
(1 e cos E
0
)(1 e cos E
f
)

1 e
2
(1 e cos E
0
)(1 e cos E
f
)
(12)
Multiplying by and rearranging gives:

t =
3
c
3
+

a
3
_
sin E
0
(e cos E
f
) sin E
f
(e cos E
0
)
(1 e cos E
0
)(1 e cos E
f
)
_
(1 e cos E
0
)(1 e cos E
f
)

1 e
2

1 e
2
(13)
Now, gather the terms in Equation 13 into terms consistent with Equations 10 and 11:

t =
3
c
3
+

a
3
_

1 e
2
sin E
0
(1 e cos E
0
)
e cos E
f
(1 e cos E
f
)

1 e
2
sin E
f
(1 e cos E
f
)
e cos E
0
(1 e cos E
0
)
_
, (14)
where
=
(1 e cos E
0
)(1 e cos E
f
)

1 e
2
. (15)
Write Equation 14 in terms of sin and cos :

t =
3
c
3
+

a
3
_
sin
0
(cos
f
) sin
f
(cos
0
)
_
_
(1 e cos E
0
)(1 e cos E
f
)

1 e
2
_
. (16)
Use the trig identity given in Equation 8 and the fact that

a
4
/

a =

a
3
to obtain:

t =
3
c
3
+ sin()
a(1 e cos E
0
)a(1 e cos E
f
)
_
a(1 e
2
)
. (17)
The magnitude of the position vector can be expressed in terms of a, e, and E as follows:
r = a(1 e cos E). (18)
3
Therefore,

t =
3
c
3
+
r
0
r
f
sin
_
a(1 e
2
)

1 cos

1 cos
(19)

t =
3
c
3
+

r
0
r
f
sin

1 cos

r
0
r
f

1 cos
_
a(1 e
2
)
. (20)
Let the variables A and y be dened as:
A =

r
0
r
f
sin

1 cos
(21)
y =
r
0
r
f
(1 cos )
a (1 e
2
)
. (22)
and Equation 20 may be written as

t =
3
c
3
+ A

y. (23)
In terms of the time of ight:
t =

3
c
3
+ A

. (24)
This is the universal variable form that can be found in Bate, Mueller, & White, and Vallado. Many of the
variables can also be expressed equivalently using the variable c
2
given in Equation 4:
=
_
y
c
2
(25)
A = DM
_
r
0
r
f
(1 + cos ) (26)
y = r
0
+ r
f
+
A
_
E
2
c
3
1
_

c
2
, (27)
where
DM = Direction of Motion =
_
+1 if <
-1 if > .
Many texts also replace the quantity E
2
with ( = E
2
).
A simple way to derive these equation in terms of is to use the f and g expressions. The derivations
for f and g may be found in Bate, Mueller, & White (see Chapter 4). Then, one can iterate on until the
desired time of ight is obtained. Vallado uses a secant method. The method is presented in 7.6 (pp. 463)
and is also given in the next section. The solution to Lamberts problem then is an iterative process on ,
where the variables , c
2
, c
3
, and y are computed for each new value of . The process continues until the
desired time of ight is achieved. Note that the value of A is only a function of the magnitude of the two
positions and the change in the true anomaly, and is not updated.
4
ALGORITHM
Input
r
0
= Initial Position Vector
r
f
= Final Position Vector
t
0
= Desired Transfer Time
DM = Direction of Motion (1 or +1)
Initialize
cos =
r
0
r
f
|r
0
r
f
|
A = DM
_
r
0
r
f
(1 + cos )
IF = 0, A = 0
Trajectory cant be computed.
END IF
= 0
c
2
=
1
2
, c
3
=
1
6

up
= 4
2

low
= 4
WHILE |t t
0
| > 1 10
6
y = r
0
+ r
f
+
A(c
3
1)

c
2
IF (A > 0.0 & y < 0.0)
Readjust
low
until y > 0.0
END IF
=
_
y
c
2
t =

3
c
3
+ A

IF (t t
0
)

low
=
ELSE

up
=
END IF
5
=

up
+
low
2
IF ( > 1 10
6
)
c
2
=
1.0 cos

c
3
=

sin

3
ELSE IF ( < 1 10
6
)
c
2
=
1.0 cosh

c
3
=
sinh

_
()
3
ELSE
c
2
=
1
2
c
3
=
1
6
END IF
END WHILE
Compute
f = 1
y
r
0
g = 1
y
r
f
g = A
_
y

Output
v
0
=
r
f
fr
0
g
v
f
=
g r
f
r
0
g
REFERENCES
[1] R. R. Bate, D. D. Mueller, and J. E. White, Fundamentals of Astrodynamics. Dover Publications, Inc., New York,
1971.
[2] D. A. Vallado, Fundamentals of Astrodynamics and Applications. McGraw-Hill Companies, Inc., 1997.
6