05.08.2013 Views

guidance, flight mechanics and trajectory optimization

guidance, flight mechanics and trajectory optimization

guidance, flight mechanics and trajectory optimization

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

where 2((t) is given by the first equation of (4.38). This miss is to be<br />

reduced to zero by an initial coast period of to seconds with<br />

An earlier paper by Kidd <strong>and</strong> Soule (Reference 4.18) also discusses<br />

optimum rendezvous in a gravity free space. The basis of the <strong>optimization</strong><br />

is that the optimum initial conditions for a rendezvous maneuver have the<br />

relative velocity vector oriented along the line of sight. (This conclusion<br />

can be verified by examination of Figure 4.2 which shows that for constant<br />

values of the other parameters, a minimum tf occurs when yo=O.) If it is<br />

assumed that midcourse corrections have placed the rendezvous vehicle on a<br />

<strong>trajectory</strong> such that the relative velocity vector is nearly parallel to the<br />

line of sight, then the thrust vector will make only small angles with the LOS.<br />

In this paper, the cost function was selected to be the amount of energy<br />

expended normal to the line of sight,(i.e., the integral of the impulse due<br />

to the non-zero angle between the LOS <strong>and</strong> the thrust vector) <strong>and</strong> the<br />

<strong>optimization</strong> was performed by a variational technique. The range of initial<br />

.conditions over which the results are valid are limited by the same<br />

assumption as the first method of this section with the additional re-<br />

striction that the initial relative velocity vector must be nearly parallel<br />

to the LOS. This method seems to have no advantage over Gunckel's method<br />

<strong>and</strong>, therefore, will not be discussed further here.<br />

2.4.3.2 Linear Gravity Model<br />

The analysis in Section 2.4.1 is limited as an on-board mechanization<br />

because of the necessity of using iterative methods to achieve a solution.<br />

However, a method discussed in Reference (4.18) approximates an optimum<br />

policy for the terminal maneuver; although an iterative solution is still<br />

required, the convergence was found, by the authors of (L+.18), to be rapid<br />

for a wide range of initial conditions. The basis of the approximation is<br />

the observation that for a range of trajectories, there are only small<br />

variations in the thrust angle ! d" I. This observation suggests an ex-<br />

pansion for $ of the form<br />

where $ >7 (a t& tcftr))<br />

(4.41)<br />

The procedure assumes that this is the form of the optimal steering angle,<br />

then inserts this form in the equations of motion <strong>and</strong> finally determines the<br />

constants so that rendezvous actually occurs. Assuming a circular target<br />

orbit,(the equations of relative motion are derived in Section 2.1.3.2) <strong>and</strong><br />

neglecting the out of plane motion, these equations are<br />

80

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!