JimB
Here
is Ray Hudson's last input. He's at the email below if you'd like to ask
him any specific questions. Tom
Cassidy ( wegerb@gat.com
) is president and CEO (General Atomics Aeronautical Systems
Inc) Their
web-site is at http://www.gat.com/asi/aero.html
His corporation
makes most, if not all of the US drones/RPV's presently in service
and under construction. He
has no reservations at all about the technical feasibility or development
risk of such a system (as RoboLander).
I am continually
updating that RoboLander site at http://www.iasa.com.au/folders/RoboLander_files/RoboLander.htm
(and its links) as I get queries or input. regards
JS
-----Original
Message-----
From: Raymond Hudson [mailto: Rainman@tree-o-life.org
]
Sent: Saturday, 22 September 2001 12:44 PM
To: safety@iasa-intl.com; fairburn_reid@si.com
Cc: rldunham@compuserve.com; TELLFAA@FAA.gov; Jim Wes
Subject: Re: [bluecoat] Re: Remote Piloted Vehicles and Exclusionary
WPTs
I would have to disagree with the answers
to Q1 and Q2 provided below:
Q1.....No action on equipping
airplanes with remote guidance. This has
been
considered but in light of the extensive C/B panel, anyone could disable
the
airplane systems required to effect the remote pilot connection.
This kind of
rules out remote control. Other actions
are more reasonable at the current
time. However, new airplanes could build a feature like
this into the design
and it is feasible with current technology. <<Quite
simply any CB's for a RoboLander system should be located in the on-the-ground
(only) access panel (but duplicated in the E&E bay) and the
system power should be via an auxiliary load-centre, accessible inflight
via the E&E hatch (which would be code-locked / combination held
only by the captain. He then would be in the same situation
as regards his personalised ROBO squawk. Only he would know it
and only he could disclose it.). The concept therefore remains
fail-safe.In
the future, once wiring bundles are isolated, remoted or inaccessible
and circuit-breakers are physically inside the code-locked E&E bay,
(being replaced in the cockpit by status lights) then would-be hijackers
will be unable to down the aircraft. However the RoboLander system
is presently designed only to stop normal hijackings and suicide
terrorists who might otherwise repeat their 11 Sep routine. Unless systems
were physically protected from their predations, hijackers operating
with impunity could always "down" an aircraft. This may not
be the case in future designs (hopefully).>>
Ray Hudson Answer:
Q1) The other
(green) answer to this question has
the right idea. All airplanes with CAT III have a minimum
of dual autopilots (and virtually
all large airplanes have 2 even if not CAT III). One of these
A/Ps could have all power
sources reassigned to C/Bs on such an inaccessible panel when
in-air. Much like specific instruments
are required to be on an "essential" power bus that
can power it with all gens failed (powered by
batt and static inverters). Another solution which would
leave all autoflight (or robo-flight) :-) breakers
still available to crew ( but
not to hijackers )
would simply be to build a C/B lock-box for those critical breakers.
Many, many solutions to this problem,
and many not as expensive as you might think.
Q2.....Could not be controlled with today's airplanes since
the pilot can fly it
anywhere he wants manually and that is the way it almost has to be.
One of the big considerations on systems like remote control is the
importance
of not taking the pilots prerogatives
away when not desired, and not limiting his
authority unnecessarily.
Ray Hudson Answer:
Q2) The answer
provided to this one does consider the dual-tandem operation of
virtually all Electro-Hydraulic
Valve (EHV) control surface actuators used by autopilots (and
a similar parallel redundancy for
"cable grabber" autopilots that use electric motors).
Autopilot EHV's always receive their hydraulic power
from a DIFFERENT hydraulic source than the pilot's control wheel,
mechanically-operated hydraulic
valve. This is per-design so that a cable failure to the surface
will still allow the A/P to fly the airplane
under this failure condition (a parallel control path).
This feature could be exploited in such
a way to rob control from hijackers in the cockpit. The system
that reconfigured the aircraft to make
the cockpit "go dark"
could close shutoff valves to the mechanical input
valves to the control surfaces, thereby only leaving the autopilot
EHVs with hydraulic power. It becomes
even easier when the airplane is full fly-by-wire (a la Airbus family
and 777) since you can simply
ignore the pilot's wheel (or Airbus
side-stick) electronic inputs to the flight control system.
I repeat from earlier EMAILs of mine:
Such a system is WELL within our technology, even on older airplanes,
and the operational issues are not insurmountable. I try not to be an
arrogant person, but I am very much
an expert in flight control system design. I may not know much
about anything else in this world
(and will readily admit it) but airplanes and autopilot design
is certainly not one of them. >One of the big considerations
on systems like remote control is the importance
>of not taking the pilots prerogatives away when not desired, and
not limiting his
>authority unnecessarily. And
to this I would add an observation I have made before. I design systems
full-well knowing what their failure
modes are. And we design them to fail in specific ways because
we know the PILOT is the strongest
link for safety when it comes to equipment malfunction or external weather
phenomenon that may endanger a flight.
However, in the situation of a hijacking, that strongest link
immediately becomes one of the weakest links, for a hijacker only need
to usurp the pilot's authority (by
either killing them, or appealing to their surrender by killing others
onboard). Taking away control from
ANY person onboard (only in such situations as hijacking) spells
"game over" for both the hijackers
or any flight crew coerced by the hijackers. It removes the
option. Of course, the safety
and reliability of the system during normal (non-hijack and failure
conditions) would need to be assured thru proper design, just like the
Autoland systems are designed and
certified. Design criteria for such "hijacker denial"
systems could be written by industry
in the form of an FAA Advisory Circular (and I think the industry
should begin work on such an AC). Furthermore,
ground-based control is only one option which needs to be considered
and evaluated. The other is simple
on-board autoflight and the "restricted WPT" feature in the
FMS. Couple the secured AP to a secured
FMC, and the airplane will fly itself to the ground, with minimal
need for interaction by ground or ATC.
But a secure data-link to provide ATC some "inputs"
to how the airplane is flown to safety certainly
would be greater safety redundancy for the end-item
system. This should be done,
it should be studied, and a Proof Of Concept installation on a non-fly-by
wire airplane would certainly....well....
prove the concept! :-) Kind
regards, Ray
Hudson
----- Original Message -----
Sent: Friday, 21 September, 2001
14:14
Subject: [bluecoat] Re: Remote
Piloted Vehicles and Exclusionary WPTs
Other
Solutions (and answers in green/<<chevrons>>
to the comments below).
http://www.iasa-tl.com/folders/RoboLander_files/RoboLander.htm
Basic
concept http://www.iasa.com.au/folders/RoboLander_files/Behind_Closed_Doors.htm
Justifications
http://www.iasa.com.au/folders/RoboLander_files/robofaq.html
Ssome
Comments and faq http://www.iasa.com.au/folders/RoboLander_files/RoboLander1.html#7yrs
GPS
based precision landing system http://www.iasa.com.au/folders/RoboLander_files/alternatives_to_cat_iii_ils_auto.htm
DGPS
accuracy [as might be applied to such a system]
____________________Reply Separator____________________
Subject: [bluecoat] Remote Piloted Vehicles and
Exclusionary WPTs
Author: Daire97@aol.com
Date: 9/17/2001 10:07 PM
Dear Blues in light of recent events I must direct this question to
my
engineering friends at Smith Industies, BMAC Northrop/Grumman and
GE.
Q1 What has been accomplished in the field of RPV pertaining
to adaptation
to (retrofit) existing commercial flight guidance and control architecture.
Q2 Would you consider an exclusionary WPT table
within a WPT database that
would prevent navigation to or through prohibited airspace be it engaged
in autopilot or manual?
http://www.pprune.org/cgibin/ultimatebb.cgi?ubb=get_topic&f=1&t=015692
| Author |
Topic:
FAA Looking for Bright Ideas |
RATBOY
Just another
number
Member # 21333 |
posted
21 September 2001 18:25
U.S. FAA
is looking for bright ideas in dealing with the security
of the aviation system. They have set up an EMAIL address
TELLFAA@FAA.gov and a fax telephone number 202-267-5091.
This is for all areas that FAA can do anything about (operations,
procedures, technology) not other things that may be bothering
one.
Posts: 163
| From: USA | Registered: Sep 2000 |
IP: Logged |
|
|