At 04:06 PM 10/29/2001, Russ Paielli wrote:

Dear Prof. Jackson,

As you may recall, I missed your visit to Ames back in August because I
was on vacation in Michigan.  However, as the TSAFE project leader, I am
obviously very interested in your work.  I probably should have
contacted you sooner.  In any case, please keep me "in the loop" on your
TSAFE plans and progress.

For my planning purposes, perhaps you would give a rough idea of what
you intend to produce and the time frame.  I am not looking for anything
in detail that would take more than a few minutes to jot down.  If you
sent something like this to Heinz already, please let me know (his
office is right next to mine, but he is not in at the moment).

Regards,
Russ Paielli

========================================================================

Daniel Jackson wrote:

russ,

we were sorry to have missed you, but we're very much looking forward to
working with you.

here are our current plans:

1. clarify terminology and the basic requirements of TSAFE. we've made some
progress doing this, and i'll be sending you a short document in a
subsequent message.

2. build some models and simulations to understand the requirements of
TSAFE. we have started this, and plan to do it for about a month.

3. develop a design for TSAFE in an object-oriented, design pattern
paradigm. we expect to complete this in the early spring.

4. build a prototype implementation, and perhaps test it using the ETMS
feed we have here.

5. attempt to show that the implementation conforms to the design by
various forms of analysis that we are working on in our research.

our hope is that the design we produce may influence your work. we're eager
to help in any way we can, so let's have a phone conversation sometime soon
about how we might help you. how's thursday nov 1, 11am your time?

/daniel

========================================================================

From: Russ Paielli <rpaielli@mail.arc.nasa.gov>
Date: Tue, 30 Oct 2001 21:37:42 -0500
To: tsafe@geyer.lcs.mit.edu
Subject: Re: TSAFE

Prof. Jackson,

Thanks for the prompt reply.  It has exactly the kind of information I was
hoping for.  As for the phone call, you're welcome to call me anytime at
650-604-5454.  If you'd like to get Heinz in on it, he is unavailable at the
time you suggested.  Perhaps we can make it for 11:00 am or later on Friday
instead.  Let me know what's good for you.

Now I'll take a crack at the questions in the PDF document you sent.

1.  Vertical intent always comes from the air traffic controllers.  They
decide which altitude every airplane should go to, and they always enter that
altitude into the host computer, I believe.  The rate at which the airplanes
change altitude, however, is up to the pilot.  Each aircraft has a certain
reasonable range of climb and descent rates.  CTAS currently uses a single
nominal rate, but TSAFE will have to account for a range of possible rates
(but for a shorter period of time, ~3 minutes).

2.  Yes, heading is the ground track angle.  The orientation of the aircraft
about the vertical axis is called yaw.  If you really want to get technical,
heading is the direction of the velocity vector relative to the (possibly
moving) air mass, whereas course is its direction relative to the ground.
However, the term heading is often used loosely to mean course.

3.  The reference heading is simply the nominal heading that the aircraft is
supposed to be flying, regardless of its cross-track position, I think.  CTAS
currently uses the cross-track position error and the heading error to
determine if an airplane is "on track."  I've been thinking about a way to
combine the two into a more general test.  You can use Heinz's criteria for
now, but it could change later.

4.  Yes, we want to detect incursions into restricted airspace.  We also want
to detect incursions into another sector without a proper handoff.  As for
hazardous weather, I'm not sure about that one yet.  It seems that areas of
severe weather can eventually be modeled the same as restricted airspace, so
maybe we don't need to worry about that for now.

5. Perhaps we can better discuss this one on the phone.  Let me just make one
comment on the diagrams.  I have difficulty distinguishing the dotted lines
and boxes from the solid ones.  I suggest you use a more distinguishable line
style.

Russ