[From Bill Powers (961017.1020 MDT)]
Hans Blom, 961017b --
The firings continued until the closest predicted approach to Mir
was where it was intended to be. So the _present_ thruster firings
were adjusted until the _presently predicted_ position of the
shuttle was in its reference condition.
I would say this differently: The ultimate goal is the linkup. There
is not yet a reference for _how_ to get there; that could be done in
a great many different ways. Yet it has to be done in one specific
way.
In this case, the system was explained in some detail. This was an
"minus-r-bar" approach in which the shuttle started its final approach from
a station-keeping position roughly 400 feet below the Mir (the minus-r-bar
approach takes advantage of orbital mechanics to make it unnecessary ever to
fire a thrustor in the direction of the Mir and its delicate solar panels).
Then it was to make an approach directly up the radius vector to a
stationary position 30 feet beneath the docking module. During this
transition, the reference condition was known throughout, and the predicted
position of closest approach was maintained at that 30-foot position while
approach velocity was kept at a fraction of a foot per second -- one
present-time controlled variable, one predicted controlled variable. Once
that position was reached, the automatic predictions were discontinued, and
the pilot completed the approach to the shuttle by manual control and
present-time visual monitoring of relative position supplemented by
verbally-reported readings from a laser range-finder. Over that short
distance the orbital-mechanics effects were negligible.
After the separation, the controlled variable became a continuous motion in
a circle going around the Mir at a distance of about 500 feet. The
predictive system was again used to maintain this path for 1-1/2 revolutions
around the Mir.
... given the end goal, take
the most appropriate step (according to some criterium, usually
minimal expected error given the limitations of the actuators) in
that direction. Is it going to get you there? You don't know. But
it's the best you can do.
Doesn't that fit nicely into the HPCT scheme of things?
Sure. That's part of the story. There's also goal _maintenance,_ which is
done once you get to the final state. And, as in the case of the Shuttle,
trajectory control, where you maintain a specific path or set of derivatives
at specified levels while the approach in taking place.
Best,
Bill P.