Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Who's At Fault in UAV/Part91 MAC?

5 views
Skip to first unread message

Larry Dighera

unread,
Apr 22, 2004, 11:22:25 AM4/22/04
to

How does the military's use of Unmanned Aerial Vehicle intend to
comply with the Part 91 See-And-Avoid mandate? Will there be new
Restricted Areas imposed along the border, or will the UAVs be flown
in Positive Control Airspace?

-------------------------------------------------------------------
AVflash Volume 10, Number 17b -- April 22, 2004

-------------------------------------------------------------------

...BORDER SECURITY BY DRONE
Chances are you won't have to join the military to encounter a
UAV. They've been in limited use over U.S. airspace for years but
the Department of Homeland Security wants to use them regularly to
patrol the border between Arizona and Mexico. The San Diego
Union-Tribune reports that the flights are expected to start later
this month and the U.S. Customs and Border Patrol is confident
they'll help stem the tide of illegal immigrants and drugs. "It's
deal terrain," commissioner Robert Bonner told the Union-Tribune.
"There's nothing to hide. Not a tree in sight."
http://www.avweb.com/eletter/archives/avflash/226-full.html#187152


While I'm in favor of policing the nation's borders, I'm very
suspicious of ill conceived DHS measures that create hazards and
inconvenience due to the imposition of less than enlightened
practices.

The Honorable Robert Bonner
Commissioner
U.S. Customs and Border Protection
1300 Pennsylvania Ave., N.W.
Room 3.4A
Washington, D.C. 20229 Phone: (202) 927-8727
Fax: (202) 927-1393


--------------------------------

http://www.signonsandiego.com/news/military/20040419-9999-1n19uav.html
The military usually operates robotic surveillance aircraft in
restricted areas throughout the United States.

But efforts to broaden the use of robotic aircraft in civilian
airspace have furrowed some brows in the aviation community, and
one aviation safety group opposes it.

The propeller-driven Predator resembles a small airplane with no
cockpit. The plane is directed by a pilot sitting in a ground
control station that receives live images transmitted by cameras
in the aircraft's spoon-shaped nose. It was designed to remain
aloft for 40 hours at altitudes as high as 25,000 feet.

San Diego-based General Atomics Aeronautical Systems, which makes
the Predator and other unmanned aerial vehicles, or UAVs, won't
discuss the specific work it does for government customers. But
spokeswoman Cyndi Wegerbauer said the company is getting more
requests to fly UAVs along the U.S. coast and borders today than
in the past.

"We have done work along the borders now with Predator, Predator B
and even I-GNAT, so it's not new to us," said Wegerbauer,
referring to the company's UAV models. "But the acceptability of
using these systems for border surveillance has increased
dramatically since terrorism became such a real, in-our-back yard
threat."

Such requests are driving the demand to fly UAVs in civilian
airspace, experts said.

"We're on the threshold of a new era in terms of using UAVs in
civil and commercial airspace in the United States," said Scott
Dan, who oversees UAV research and development at GA Aeronautical
Systems. He sees UAVs being used someday to fly over wildfires and
perhaps even for agricultural purposes such as crop-dusting.

Dan has been working to realize that vision by serving as
president of the UAV National Industry Team, which is developing
the regulations and technology required to make it easier to fly
UAVs in civilian airspace. Participating companies include GA
Aeronautical Systems, Lockheed Martin, Boeing, Northrop Grumman,
Aerovironment and Aurora Flight Sciences.

NASA has provided $101 million to fund Access 5, whose goal is to
ease rules for UAV flights in civilian airspace in five years.

As part of the process, the two groups also are working on
procedures for autonomous UAVs, such as the Global Hawk developed
in San Diego by Northrop Grumman. An autonomous UAV follows a
computer-programmed route.

UAVs are permitted to fly in civilian airspace under a certificate
of authorization granted by the Federal Aviation Administration.
It requires the operator to file a flight plan at least 30 days in
advance.

The goal of UAV proponents is to allow them to fly routinely into
and out of designated U.S. airports. Operators would be allowed to
file a flight plan and fly on the same day, just like any pilot.

To the National Air Disaster Alliance, a flight safety group in
Washington, D.C., the idea of routinely flying remote-controlled
airplanes in civilian airspace is folly.

"Unfortunately for the American public, this is not an issue that
has appeared on the radar," said Tom O'Mara, an alliance board
member. "We already have a problem with air traffic control. Our
skies are overcrowded as it is. So why would anyone want to put an
unmanned aircraft into that mix? It's just a bad idea."

The issue is "problematic" for pilots, said John Mazor, a
spokesman for the Air Line Pilots Association.

"If it's implemented properly, you might look at it and think, 'It
looks like a good idea.' But the devil is in the details," Mazor
said. "We don't want anything flying around that would reduce the
safety and separation requirements for commercial airliners."

For Jeff Myers of the Aircraft Owners and Pilots Association, a
general aviation group, one concern stems from the stealthy nature
of UAV operations by the military and the Department of Homeland
Security.

"As long as they're in military airspace, there's no problem
because they're operating on their own side of the fence," Myers
said. "In this case, the fence is a matter of communications."

The FAA has no authority over UAV flights in military airspace,
said William Shumann, an agency spokesman in Washington, D.C.

"Currently, there are no FAA regulations dealing with the
certification of UAV pilots, aircraft or (commercial) operators,"
he said.

But using the FAA's certificates of authorization, companies such
as GA Aeronautical Systems fly UAVs in civilian airspace.

In a Dec. 16 letter to Unmanned Systems magazine, GA Aeronautical
Systems President Thomas J. Cassidy said UAVs are treated by air
traffic controllers like any other small aircraft "because that is
what they are."

For example, Cassidy said a Predator B was flying at an altitude
of 21,000 feet last year in eastern California when "numerous
airliners in the vicinity" were calling air traffic controllers.

The pilots were asking for the best altitudes to avoid turbulence,
Cassidy said. A controller called by radio to the Predator, which
relayed the signal to the pilot in the ground control station.

"The pilot of Predator B 002, who was located hundreds of miles
from the aircraft, responded with a 'smooth ride at FL 210,' "
Cassidy wrote, referring to a flight level of 21,000 feet. "As far
as the controller was concerned, the Predator B was just another
airplane."

--------------------------------

http://www.azcentral.com/specials/special03/articles/0331customs-ON.html

President Bush has requested $64 million to develop and obtain
more technology to assist with border protection. Another $10
million has been requested to develop and deploy unmanned aerial
vehicles, essentially drones, that would look out for illegal
border crossers.

By late spring or early summer, the agency hopes to deploy some of
the drones along the Arizona border, where arrests have risen
recently.

-----------------------------------
http://appropriations.house.gov/_files/RobertBonnerTestimony.pdf

Unmanned Aerial Vehicles (UAVs)

Like ISIS, Unmanned Aerial Vehicles (UAVs) are both an important
part of the smarter border strategy and an essential element of
the Border Patrol’s revised National Strategy. UAVs equipped with
sophisticated on-board sensors have the potential to provide
unparalleled surveillance capability. UAVs provide long-range
surveillance. As a result, they are especially effective
force-multipliers because they have the capacity to remain on
station much longer than other airborne assets, and are
particularly useful for monitoring remote land border areas where
patrols cannot easily travel and infrastructure is difficult or
impossible to build. UAVs will perform missions involving
gathering intelligence on border activities was well as conducting
surveillance over open water along the Gulf Coast, the Florida
peninsula and the Great Lakes region on the northern border. The
high endurance of the larger classes of UAVs permits
uninterrupted overnight or around-the-clock coverage, and the size
and operating altitudes can make UAVs effectively undetectable
by unaided human senses. UAVs will also contribute to enforcement
effectiveness and officer safety by providing communications links
for coordinating multiple units on the ground is important in
remote border operating areas. The $10 million in funding sought
for UAVs will enable CBP to capitalize more fully on the UAV
research that has taken place in a military context, and to apply
UAVs in support of the Homeland Security mission. The funding
would allow CBP to deploy and operate a system of unmanned aerial
vehicles in support of the Border Patrol and other components of
Customs and Border Protection. The use of UAVs will complement the
other intrusion detection and intelligence gathering components of
the border surveillance network to meet the mission of stopping
the illegal entry of terrorists, smugglers and others into the
United States.

-----------------------------------
http://uav.navair.navy.mil/airdemo03/articles03/fednews.htm

“Let’s say you’re the chief of a Border Patrol sector, and it
takes six to seven agents to fly a UAV,” Thrash said. “You have to
make the choice: ‘Is the UAV providing me enough surveillance
capability to keep six or seven agents off the line?’”

...

However, obstacles remain before UAVs can fly along the borders.
First, Homeland Security has to secure permission from the Federal
Aviation Administration to fly the unmanned systems in commercial
airspace.

Another issue is a concern over citizen privacy.

The federal government is able to conduct surveillance using
unmanned systems that exceeded the imagination only a few years
ago, said Barry Steinhardt, director of the American Civil
Liberties Union technology and liberty program. Very few laws
govern when and how such technologies can be used, Steinhardt
said.

“We’re creating this surveillance monster — the planes are just
one example of that — and we’re creating it at light speed,”
Steinhardt said.

-------------------------------------
http://www.upi.com/view.cfm?StoryID=20030616-093730-1766r
In the field test, King buried two of his new sensors in the
ground so their devices' antennas stuck up a few inches. He
measured each one's precise latitude and longitude with a Global
Positioning System gadget and entered the coordinates in his
laptop.

While the Border Hawk circled a couple of hundred feet overhead,
buzzing like a large mosquito, four APB members and myself walked
past the hidden motion detectors single file. ("SBIs always walk
single file," I was told.) The two gizmos successfully reported by
radio our direction and speed, although they overestimated our
numbers, signaling that there were 11 of us instead of five.

Our GPS coordinates showed up on a map on King's
wireless-networked laptop and a volunteer, who is a model airplane
hobbyist, piloted the Border Hawk to our location to record our
presence. Somebody who happened to be logged onto ABP's Web site
at that moment could have watched live aerial pictures of me
squinting up at the drone.

-------------------------------------

http://www.space.com/businesstechnology/technology/ufos_vs_uavs_030813.html

-------------------------------------

In my opinion, it is a very helpful (and in some instances quite
necessary) virtue to be able to take criticism even if it is offensive
or insulting. In fact, even the most offensive criticism might (and
hopefully does!) contain insights that are valuable, and by
disregarding the entire criticism, you are throwing away that
insight. You may not like it, but it sometimes does pay to listen to
a person that is not as friendly as you'd like her to be.
-- Tobias Dussa <tdu...@sdhs.de>

John T

unread,
Apr 22, 2004, 11:56:33 AM4/22/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:u7lf80tvu48jvs2vu...@4ax.com

>
> How does the military's use of Unmanned Aerial Vehicle intend to
> comply with the Part 91 See-And-Avoid mandate? Will there be new
> Restricted Areas imposed along the border, or will the UAVs be flown
> in Positive Control Airspace?

It's not just the military, but civilian government agencies that are
considering the use of UAV's.

If the UAV's are in the flight levels, then they will be in Positive Control
Airspace, right?

If the UAV's are for border patrol, would it not be reasonable to expect
them to be within a few miles of the border? As such, how much of an issue
would you expect them to be to Part 91 flights? Or are you concerned about
the occassional drug-running flight? :)

As for your question re: border restricted areas, I have to question how
many Part 91 flights are conducted close enough to the border for this to be
a problem. Do you know how many occur in any given time frame?

UAV use in general airspace should be carefully considered before
implementation, but I'm not as concerned about their use in border patrol
use as I am about their loitering over a city with several nearby airports
and busy airspace.

As for your subject line question, I'd wait for an NTSB ruling before
passing judgment on that.

--
John T
http://tknowlogy.com/TknoFlyer
http://www.pocketgear.com/products_search.asp?developerid=4415
____________________


Steven P. McNicoll

unread,
Apr 22, 2004, 12:07:41 PM4/22/04
to

"Larry Dighera" <LDig...@att.net> wrote in message
news:u7lf80tvu48jvs2vu...@4ax.com...

>
> How does the military's use of Unmanned Aerial Vehicle intend to
> comply with the Part 91 See-And-Avoid mandate? Will there be new
> Restricted Areas imposed along the border, or will the UAVs be flown
> in Positive Control Airspace?
>

Did you read the item before posting your message? There is no altitude
reference outside of Class A airspace, so presumably see-and-avoid is not an
issue. Yeah, they've got to climb through Class E airspace to get to Class
A, but I'd assume that'll be done in a restricted area.


Tony Cox

unread,
Apr 22, 2004, 12:17:37 PM4/22/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:u7lf80tvu48jvs2vu...@4ax.com...
>
> How does the military's use of Unmanned Aerial Vehicle intend to
> comply with the Part 91 See-And-Avoid mandate? Will there be new
> Restricted Areas imposed along the border, or will the UAVs be flown
> in Positive Control Airspace?

Good question. In fact, several good questions.

As for "see-and-avoid", there is some visual feedback to the remote
pilot ahead. I've no idea as to the quality of the circuit, nor the coverage
of the sky from the cockpit it gives the pilot. If the quality is sufficient
that the remote pilot can pass the aviation medical eye exam over the
remote circuit (put the eye chart in front of the plane and ask the remote
pilot what the letters say, color charts, etc.), I don't see the problem ;-)

I note that there is currently no requirement for certification, even
medical
requirements. With all the fuss over sport pilot medical requirements, this
needs to be addressed or we might as well not have medical certifications
at all..


Tony Cox

unread,
Apr 22, 2004, 12:24:34 PM4/22/04
to
"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:h5Shc.18545$l75....@newsread2.news.atl.earthlink.net...

And in case of mechanical failure when the plane has to descend
into the VFR altitudes? Don't you think the rest of us are entitled to
be "seen-and-avoided"?


Steven P. McNicoll

unread,
Apr 22, 2004, 12:57:01 PM4/22/04
to

"Tony Cox" <t...@coxrt.com> wrote in message
news:6lShc.7006$e4....@newsread2.news.pas.earthlink.net...

>
> And in case of mechanical failure when the plane has to descend
> into the VFR altitudes? Don't you think the rest of us are entitled to
> be "seen-and-avoided"?
>

A mechanical failure would make it an aircraft in distress. An aircraft in
distress has the right-of-way over all other air traffic.


Tony Cox

unread,
Apr 22, 2004, 1:13:20 PM4/22/04
to
"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:xPShc.18589$l75....@newsread2.news.atl.earthlink.net...

Indeed. But it makes it *more dangerous*, which is why Larry's
post is relevant, even if UAV's are supposedly confined to Class A
airspace.

Steven P. McNicoll

unread,
Apr 22, 2004, 1:18:57 PM4/22/04
to

"Tony Cox" <t...@coxrt.com> wrote in message
news:Q2Thc.7045$e4....@newsread2.news.pas.earthlink.net...

>
> Indeed. But it makes it *more dangerous*, which is why Larry's
> post is relevant, even if UAV's are supposedly confined to Class A
> airspace.
>

It makes it *more dangerous* only in the sense that more traffic makes
flying more dangerous. Any aircraft can have a mechanical failure that
affects it's ability to maintain altitude and/or maneuver. It's not *more
dangerous* simply because it's a UAV.


Peter Gottlieb

unread,
Apr 22, 2004, 1:20:43 PM4/22/04
to

"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:xPShc.18589$l75....@newsread2.news.atl.earthlink.net...

A lot of good that will do you when the military doesn't inform civilian
authorities and thus you have no idea an unmanned drone is rapidly
descending on you.

There is no way unmanned aircraft can mix with all others and not have some
reduction in safety. The questions are just how much of a reduction, what
can be done to mitigate the danger, and regulations which do not penalize GA
pilots for being in the wrong place at the wrong time.

I imagine these UAVs will be common in the future, maybe even extending to
pilot-less GA and commercial aircraft. How much this industry grows will in
some part be determined by it's safety record. A few bad incidents will
slow or stop progress in this direction so I would hope the companies
involved have the foresight and intelligence to do the proper engineering
and risk analysis to keep incidents from happening. I also hope their
solution to assuring safety is not a legislative one where they manage to
convince Washington to prohibit GA aircraft from operating anywhere near one
of these things.

Steven P. McNicoll

unread,
Apr 22, 2004, 1:28:03 PM4/22/04
to

"Peter Gottlieb" <ng-repl...@icq.net> wrote in message
news:L9Thc.95409$_g4.21...@news4.srv.hcvlny.cv.net...

>
> A lot of good that will do you when the military doesn't inform civilian
> authorities and thus you have no idea an unmanned drone is rapidly
> descending on you.
>

They're not drones.


>
> There is no way unmanned aircraft can mix with all others and not
> have some reduction in safety. The questions are just how much of
> a reduction, what can be done to mitigate the danger, and
> regulations which do not penalize GA pilots for being in the wrong
> place at the wrong time.
>

If operations are confined to Restricted Areas and Class A airspace they're
not mixing with all others.


Tony Cox

unread,
Apr 22, 2004, 2:07:41 PM4/22/04
to
"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:58Thc.18603$l75....@newsread2.news.atl.earthlink.net...

It is if the pilot can't scan for traffic or search for a suitable emergency
landing site.

In any case, an aircraft "in distress" is only allowed to violate the
FAR's as far as necessary to deal with the emergency. I'd have
a hard time proving reasonable violation of "See and Avoid" in the
simple case of engine failure. You're proposing that these UAV's
can simply ignore this rule because they're supposed to be in class
A all the time.


Casey Wilson

unread,
Apr 22, 2004, 2:08:41 PM4/22/04
to

"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:h5Shc.18545$l75....@newsread2.news.atl.earthlink.net...

They are not always flown in restricted areas. This test facility is
not in a restricted area.

http://www.dfrc.nasa.gov/Gallery/Photo/Altair_PredatorB/HTML/ED03-0078-1.html

...and as copied from:

http://www.defense-aerospace.com/cgi-bin/client/modele.pl?session=dae.3675963.1082656906.QIgIisOa9dUAAAFVD1c&prod=18689&modele=jdc_1

"Altair is expected to be the first UAV to meet Federal Aviation
Administration requirements to operate from conventional airports, with
piloted aircraft, in the national airspace. In addition to triple-redundant
avionics, Altair is configured with a fault-tolerant, dual-architecture
flight control system. The UAV will be equipped with an automated
collision-avoidance system and an air traffic control voice relay. The relay
allows air-traffic controllers to talk to ground-based Altair pilots through
the aircraft."

On several occasions, Joshua Control has called me with warnings about
UAVs and their chase planes orbiting around El Mirage Dry Lake. The location
is a couple miles north of KREY near Adelanto, California.


Casey Wilson

unread,
Apr 22, 2004, 2:12:20 PM4/22/04
to

"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:DgThc.18609$l75....@newsread2.news.atl.earthlink.net...

Their operations are not confined to restricted areas and Class A
airspace.

Larry Dighera

unread,
Apr 22, 2004, 2:25:12 PM4/22/04
to
On Thu, 22 Apr 2004 15:56:33 GMT, "John T" <j...@his.isp> wrote in
Message-Id: <36b0f2ceda472b1f...@news.bubbanews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:u7lf80tvu48jvs2vu...@4ax.com
>>
>> How does the military's use of Unmanned Aerial Vehicle intend to
>> comply with the Part 91 See-And-Avoid mandate? Will there be new
>> Restricted Areas imposed along the border, or will the UAVs be flown
>> in Positive Control Airspace?
>
>It's not just the military, but civilian government agencies that are
>considering the use of UAV's.

The AvFlash article mentioned the Border Patrol UAVs being operated by
the military.

>If the UAV's are in the flight levels, then they will be in Positive Control
>Airspace, right?

That might be true if they are capable of adequate surveillance
performance from >18,000' MSL, but they will have to climb to that
altitude outside Positive Control Airspace, in Joint Use airspace or
Restricted airspace, as the NAS is currently structured.

>If the UAV's are for border patrol, would it not be reasonable to expect
>them to be within a few miles of the border? As such, how much of an issue
>would you expect them to be to Part 91 flights? Or are you concerned about
>the occassional drug-running flight? :)

While the UAVs may operate within a few miles of the national
boarders, I doubt they will be based there. So it is likely they will
have to traverse Joint Use airspace en route to their stations.

>As for your question re: border restricted areas, I have to question how
>many Part 91 flights are conducted close enough to the border for this to be
>a problem. Do you know how many occur in any given time frame?

Many international Part 91 flights occur each day. To intentionally
design the NAS in such a way as to permit UAV operation at reduced
vision standards is unprofessional, unacceptable to public safety, and
negligent.

>UAV use in general airspace should be carefully considered before
>implementation, but I'm not as concerned about their use in border patrol
>use as I am about their loitering over a city with several nearby airports
>and busy airspace.

And how long do you estimate it will take for UAVs to be operating
beyond the national boarder corridors, given the national hysteria?

>As for your subject line question, I'd wait for an NTSB ruling before
>passing judgment on that.

Right. It's difficult to generalize about potential MAC
responsibility without specific facts. However, once the inevitable
MAC occurs, and the Part 91 pilot is no longer able to testify (due to
his untimely death), do you expect the team operating the UAV to
actually take responsibility for their failure to see-and-avoid? From
the past behavior of military in MACs with civil aircraft, I would
expect the military to deny all responsibility.

This begs the question, how is the UAV's conspicuity planned to be
enhanced?


Matt Johnson

unread,
Apr 22, 2004, 2:26:44 PM4/22/04
to
> A mechanical failure would make it an aircraft in distress. An aircraft
in
> distress has the right-of-way over all other air traffic.

Under the existing definition, sure--but this definition presupposes that
such an aircraft has human occupants. To my thinking, unmanned hardware
can't exactly experience "distress"; therefore, right-of-way shouldn't be an
absolute in this instance. Manned air traffic should never be jeopardized
by unmanned aircraft, irrespective of any malfunction such hardware may
experience. "Safety" always pertains to the human element, never machinery.

A related thought: rockets always have a human-controlled self-destruct
capability to protect lives and property on the ground should the vehicle
experience a loss of control. Maybe UAVs should have this cabability too.

Larry Dighera

unread,
Apr 22, 2004, 2:27:15 PM4/22/04
to
On Thu, 22 Apr 2004 16:07:41 GMT, "Steven P. McNicoll"
<ronca...@nospamearthlink.net> wrote in Message-Id:
<h5Shc.18545$l75....@newsread2.news.atl.earthlink.net>:

>
>"Larry Dighera" <LDig...@att.net> wrote in message
>news:u7lf80tvu48jvs2vu...@4ax.com...
>>
>> How does the military's use of Unmanned Aerial Vehicle intend to
>> comply with the Part 91 See-And-Avoid mandate? Will there be new
>> Restricted Areas imposed along the border, or will the UAVs be flown
>> in Positive Control Airspace?
>>
>
>Did you read the item before posting your message?

Yes. Did you read the entire content of my article? You'll note in
one of the excerpts of the links, that the demonstration UAV was at
200' AGL.

>There is no altitude reference outside of Class A airspace, so presumably
>see-and-avoid is not an issue.

Agreed; your assessment is presumptuous.

>Yeah, they've got to climb through Class E airspace to get to Class
>A, but I'd assume that'll be done in a restricted area.

From the links I provided in the article, it seems that the FARs are
being revised to accommodate UAVs outside of restricted airspace.

Larry Dighera

unread,
Apr 22, 2004, 2:41:55 PM4/22/04
to
On Thu, 22 Apr 2004 16:17:37 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <BeShc.7002$e4....@newsread2.news.pas.earthlink.net>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:u7lf80tvu48jvs2vu...@4ax.com...
>>
>> How does the military's use of Unmanned Aerial Vehicle intend to
>> comply with the Part 91 See-And-Avoid mandate? Will there be new
>> Restricted Areas imposed along the border, or will the UAVs be flown
>> in Positive Control Airspace?
>
>Good question. In fact, several good questions.
>
>As for "see-and-avoid", there is some visual feedback to the remote
>pilot ahead. I've no idea as to the quality of the circuit, nor the coverage
>of the sky from the cockpit it gives the pilot. If the quality is sufficient
>that the remote pilot can pass the aviation medical eye exam over the
>remote circuit (put the eye chart in front of the plane and ask the remote
>pilot what the letters say, color charts, etc.), I don't see the problem ;-)

That would be a reasonable test of the UAV pilots' ability to comply
with the see-and-avoid regulation, but the UAV pilots would have to be
able to see above and below and to the sides in addition to airspace
immediately ahead.

>I note that there is currently no requirement for certification, even

>medical requirements [for UAV operators].

Can you provide a citation that supports that statement? It is scary
beyond belief if true. Imagine the uncertified pilot of the UAV
safely on the ground simultaneously monitoring video from the front,
above, below and to the sides while attempting to spot intruders on
the ground. How much time is going to be devoted to traffic scan
compared to ground scan? Will the operators receive recognition for
avoiding collisions or spotting illegals? How will the public be
assured that their priority is safety, and not mission success as is
inherent in manned aircraft where the pilots have their lives on the
line in avoiding collisions?

John T

unread,
Apr 22, 2004, 2:55:26 PM4/22/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:pnvf80lr0qav4lgae...@4ax.com

>
> The AvFlash article mentioned the Border Patrol UAVs being operated by
> the military.

I didn't say the military wouldn't be involved, but you explicitly ignored
the inclusion of non-military agencies using UAV's.

> That might be true if they are capable of adequate surveillance
> performance from >18,000' MSL,

Safe to assume. :)

> ...but they will have to climb to that


> altitude outside Positive Control Airspace, in Joint Use airspace or
> Restricted airspace, as the NAS is currently structured.

What's the problem if it's restricted space?

> While the UAVs may operate within a few miles of the national
> boarders, I doubt they will be based there. So it is likely they will
> have to traverse Joint Use airspace en route to their stations.

Perhaps. Perhaps not. UAV's don't necessarily need the massive runways
other recon aircraft require.

>> Do you know how many occur in any given
>> time frame?
>
> Many international Part 91 flights occur each day.

So the answer to my yes/no question would be...? No, you don't know.

> To intentionally
> design the NAS in such a way as to permit UAV operation at reduced
> vision standards is unprofessional, unacceptable to public safety, and
> negligent.

Unprofessional? Negligent? Reduced vision standards? What reduced
standards?

> And how long do you estimate it will take for UAVs to be operating
> beyond the national boarder corridors, given the national hysteria?

I make no assumptions - including one regarding "hysteria". The only
hysterical one here appears to be you. :)

> ...do you expect the team operating the UAV to


> actually take responsibility for their failure to see-and-avoid?

You're assuming facts no in evidence.

> From
> the past behavior of military in MACs with civil aircraft, I would
> expect the military to deny all responsibility.

Perhaps, but the NTSB would still make their ruling, wouldn't they?

> This begs the question, how is the UAV's conspicuity planned to be
> enhanced?

Has anybody said this enhancement would be made?

Tony Cox

unread,
Apr 22, 2004, 3:03:52 PM4/22/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:fs0g8057k1oo1engv...@4ax.com...

> On Thu, 22 Apr 2004 16:17:37 GMT, "Tony Cox" <t...@coxrt.com> wrote in
> Message-Id: <BeShc.7002$e4....@newsread2.news.pas.earthlink.net>:
>
> >I note that there is currently no requirement for certification, even
> >medical requirements [for UAV operators].
>
> Can you provide a citation that supports that statement?

It's a quote in your original post, attributed to one William
Shumann:- "Currently, there are no FAA regulations dealing


with the certification of UAV pilots, aircraft or (commercial)
operators," he said.

> It is scary
> beyond belief if true. Imagine the uncertified pilot of the UAV
> safely on the ground simultaneously monitoring video from the front,
> above, below and to the sides while attempting to spot intruders on
> the ground. How much time is going to be devoted to traffic scan
> compared to ground scan? Will the operators receive recognition for
> avoiding collisions or spotting illegals? How will the public be
> assured that their priority is safety, and not mission success as is
> inherent in manned aircraft where the pilots have their lives on the
> line in avoiding collisions?

What assurance do we have that he won't have a heart attack, or
loose consciousness, or a whatever?

I'm of the opinion that physically being in the plane sharpens your
mind up. When I fly, I'm constantly "on edge" and ready to react
instantly to any problem. It's my bum on the line too. Frankly, I'd
never expect that level of alertness from a remote pilot, slouched in
a chair drinking his coffee, thumbing through "Playboy" during the
dull bits of a mission, scratching his butt and wandering off to
the bathroom whenever he feels like it. All he risks is his job.


Stan Gosnell

unread,
Apr 22, 2004, 6:31:25 PM4/22/04
to
"John T" <j...@his.isp> wrote in
news:3ea7650b202d0388...@news.bubbanews.com:

> What's the problem if it's restricted space?

None, if it's restricted airspace. But it may very well be in joint use
airspace, especially if the other civilian players get into the game.
They'll be climbing through the Cessnas flying around.

>> Many international Part 91 flights occur each day.
>
> So the answer to my yes/no question would be...? No, you don't know.

I can't give you an exact number, but it's in the thousands. There are
thousands of daily helicopter flights to/from the Gulf of Mexico alone,
nevermind the true international flights, both airline and Part 135 and
Part 91 flights, US and other countries. My best guess is that it's in the
tens of thousands daily, counting everything.

We're giving up lots of freedoms to the government, and now we're expected
to possibly give our lives, for little or no return. The sky is falling,
the sky is falling!!!! Not I, said the little red hen.

--
Regards,

Stan

Larry Dighera

unread,
Apr 22, 2004, 8:08:41 PM4/22/04
to
On Thu, 22 Apr 2004 18:55:26 GMT, "John T" <j...@his.isp> wrote in
Message-Id: <3ea7650b202d0388...@news.bubbanews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:pnvf80lr0qav4lgae...@4ax.com

[...]


>>> Do you know how many occur in any given
>>> time frame?
>>
>> Many international Part 91 flights occur each day.
>
>So the answer to my yes/no question would be...? No, you don't know.

Implicit in your question is the notion that, because there are less
international Part 91 operations than domestic, there is no problem
compromising their safety. I do not hold that view.

>> To intentionally
>> design the NAS in such a way as to permit UAV operation at reduced
>> vision standards is unprofessional, unacceptable to public safety, and
>> negligent.
>
>Unprofessional? Negligent? Reduced vision standards? What reduced
>standards?

Are you implying that the ground based crew operating the UAV would be
able to meet the vision standards required of a certificated airman
and mandated by Code of Federal Regulations, Title 14, Volume 2,
Chapter 1, Part 91, Subpart A, § 91.113(b):

When weather conditions permit, regardless of whether an
operation is conducted under instrument flight rules or visual
flight rules, vigilance shall be maintained by each person
operating an aircraft so as to see and avoid other aircraft.

solely through the use of video equipment on-board the UAV? If not, I
would characterize the UAV pilot vision standards as reduced from
those required of certificated airmen.

>> And how long do you estimate it will take for UAVs to be operating
>> beyond the national boarder corridors, given the national hysteria?
>
>I make no assumptions - including one regarding "hysteria". The only
>hysterical one here appears to be you. :)

What has lead you to that conclusion?

>> ...do you expect the team operating the UAV to
>> actually take responsibility for their failure to see-and-avoid?
>
>You're assuming facts no in evidence.

You didn't answer the question. :-)

>> From
>> the past behavior of military in MACs with civil aircraft, I would
>> expect the military to deny all responsibility.
>
>Perhaps, but the NTSB would still make their ruling, wouldn't they?

The NTSB has shown a significant lack of impartiality in at least one
civil/military MAC case:
http://www.ntsb.gov/ntsb/brief.asp?ev_id=20001213X33340&key=2

>> This begs the question, how is the UAV's conspicuity planned to be
>> enhanced?
>
>Has anybody said this enhancement would be made?

Unfortunately, there has been no mention whatsoever of enhancing the
conspicuity of UAVs operating in Joint Use airspace in any of the
literature I have read. It would seem prudent to equip the UAV with a
bright light on the front of the UAV, so the pilot on a head-on
collision course with it might be able to see it in time to attempt to
avoid it. The UAV might also be equipped with TCAS to assist in
warning of an impending MAC.


John T

unread,
Apr 23, 2004, 1:08:42 AM4/23/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:eo5g80tl6v3anpv0g...@4ax.com

>
> Implicit in your question is the notion that, because there are less
> international Part 91 operations than domestic, there is no problem
> compromising their safety. I do not hold that view.

You're assuming a significant rise in the danger to other aircraft (*You*,
not I, separated Part 91 traffic from the rest.) I'm not yet convinced that
adding remotely piloted aircraft to a relatively rarely-travelled slice of
airspace over very sparsely populated border areas raises the danger to
pilots enough for me to be worried. Frankly, I'd give much better odds to
having an in-flight fire or engine failure than a MAC with a remotely
piloted aircraft. The Big Sky is much bigger in the border areas discussed
in your articles.

> Are you implying that the ground based crew operating the UAV would be

> able to meet the vision standards required of a certificated airman...


> solely through the use of video equipment on-board the UAV?

I implied no such thing. However, I'm curious to know why you're implying
they *wouldn't* be able to meet those requirements. Are you aware of all
the capabilities of the UAV's you're talking about? I'm not so I can't make
too many assumptions either way.

>> I make no assumptions - including one regarding "hysteria". The only
>> hysterical one here appears to be you. :)
>
> What has lead you to that conclusion?

What led you to yours? Does "Chicken Little" mean anything to you? :)

>>> ...do you expect the team operating the UAV to
>>> actually take responsibility for their failure to see-and-avoid?
>>
>> You're assuming facts no in evidence.
>
> You didn't answer the question. :-)

I have no expectation in your hypothetical scenario.

> The NTSB has shown a significant lack of impartiality in at least one
> civil/military MAC case:

The NTSB has shown a "significant lack of impartiality" in a number of other
cases, too, but they're still the closest thing we have to a standing
impartial review board that merits trust.

Mike Money

unread,
Apr 23, 2004, 5:22:56 AM4/23/04
to
The Predator is equipped with Terrain and In-flight Avoidance Systems.
They will see you before you see them.

Mike $$$

Larry Dighera

unread,
Apr 23, 2004, 10:55:49 AM4/23/04
to
On Thu, 22 Apr 2004 19:03:52 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <sGUhc.7155$e4....@newsread2.news.pas.earthlink.net>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:fs0g8057k1oo1engv...@4ax.com...
>> On Thu, 22 Apr 2004 16:17:37 GMT, "Tony Cox" <t...@coxrt.com> wrote in
>> Message-Id: <BeShc.7002$e4....@newsread2.news.pas.earthlink.net>:
>>
>> >I note that there is currently no requirement for certification, even
>> >medical requirements [for UAV operators].
>>
>> Can you provide a citation that supports that statement?
>
>It's a quote in your original post, attributed to one William
>Shumann:- "Currently, there are no FAA regulations dealing
>with the certification of UAV pilots, aircraft or (commercial)
>operators," he said.
>

Aircraft operation in the NAS by an uncertificated "pilot" would seem
to violate FARs.

>
>> It is scary
>> beyond belief if true. Imagine the uncertified pilot of the UAV
>> safely on the ground simultaneously monitoring video from the front,
>> above, below and to the sides while attempting to spot intruders on
>> the ground. How much time is going to be devoted to traffic scan
>> compared to ground scan? Will the operators receive recognition for
>> avoiding collisions or spotting illegals? How will the public be
>> assured that their priority is safety, and not mission success as is
>> inherent in manned aircraft where the pilots have their lives on the
>> line in avoiding collisions?
>
>What assurance do we have that he won't have a heart attack, or
>loose consciousness, or a whatever?

It's my understanding that it takes a team of about 7 to operate a
UAV. Perhaps that level of redundancy might mitigate the concerns you
raise. However, 7 border patrol officers on the ground might be more
effective in preventing illegal entries.

>I'm of the opinion that physically being in the plane sharpens your
>mind up. When I fly, I'm constantly "on edge" and ready to react
>instantly to any problem. It's my bum on the line too. Frankly, I'd
>never expect that level of alertness from a remote pilot, slouched in
>a chair drinking his coffee, thumbing through "Playboy" during the
>dull bits of a mission, scratching his butt and wandering off to
>the bathroom whenever he feels like it. All he risks is his job.

Those are my concerns as well.


Larry Dighera

unread,
Apr 23, 2004, 11:33:41 AM4/23/04
to
On Fri, 23 Apr 2004 05:08:42 GMT, "John T" <j...@his.isp> wrote in
Message-Id: <30a6052946274546...@news.bubbanews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:eo5g80tl6v3anpv0g...@4ax.com
>>
>> Implicit in your question is the notion that, because there are less
>> international Part 91 operations than domestic, there is no problem
>> compromising their safety. I do not hold that view.
>
>You're assuming a significant rise in the danger to other aircraft (*You*,
>not I, separated Part 91 traffic from the rest.) I'm not yet convinced that
>adding remotely piloted aircraft to a relatively rarely-travelled slice of
>airspace over very sparsely populated border areas raises the danger to
>pilots enough for me to be worried.

Intentionally compromising air safety is always a bad idea. Once the
UAV 'camel' has its nose under the tent, you can bet that you will be
sleeping with it soon, fleas and all.

>Frankly, I'd give much better odds to having an in-flight fire or engine
>failure than a MAC with a remotely piloted aircraft.

How did you arrive at that point of view. Do you have any data to
support it?

Giving odds or taking chances is an inappropriate approach to air
safety.

>The Big Sky is much bigger in the border areas discussed
>in your articles.

The "Big Sky" is a total myth. Any rational system that relies upon
chance to insure air safety is doomed to failure. I hope you're not
an FAA employee.

>> Are you implying that the ground based crew operating the UAV would be
>> able to meet the vision standards required of a certificated airman...
>> solely through the use of video equipment on-board the UAV?
>
>I implied no such thing.

You questioned my use of the term 'reduced vision standards'. That
lead me to believe that you felt that UAV operators would be held to
the same (not reduced) vision standards as certificated airmen. If
your questioning of my use of the term 'reduced vision standards' did
not imply your belief that they UAV operators would be held to the
same standards as certificated airmen, what were you implying?

:However, I'm curious to know why you're implying


>they *wouldn't* be able to meet those requirements. Are you aware of all
>the capabilities of the UAV's you're talking about? I'm not so I can't make
>too many assumptions either way.

The military has not disclosed to me all the capabilities of their
UAVs. :-) However, unless there is high-resolutin, color, binocular
vision in all quadrants, the UAV operators visual capability to see
and avoid will be substandard to that required of a certificated
airman.

>>> I make no assumptions - including one regarding "hysteria". The only
>>> hysterical one here appears to be you. :)
>>
>> What has lead you to that conclusion?
>
>What led you to yours? Does "Chicken Little" mean anything to you? :)

Your apparent lack of concern for air safety and reliance on chance
(Big Sky)for aircraft separation betrays your shallow understanding of
the issue.

>> From the past behavior of military in MACs with civil aircraft, I would
>> expect the military to deny all responsibility.

>Perhaps, but the NTSB would still make their ruling, wouldn't they?

>> The NTSB has shown a significant lack of impartiality in at least one


>> civil/military MAC case:
>
>The NTSB has shown a "significant lack of impartiality" in a number of other
>cases, too, but they're still the closest thing we have to a standing
>impartial review board that merits trust.

So you feel that a _biased_ (as opposed to _impartial_) governmental
investigative organization does not warrant reform? Comon' man,
think!

Larry Dighera

unread,
Apr 23, 2004, 11:36:43 AM4/23/04
to
On Fri, 23 Apr 2004 04:22:56 -0500, mmw...@webtv.net (Mike Money)
wrote in Message-Id: <7637-40...@storefull-3312.bay.webtv.net>:

>The Predator is equipped with Terrain and In-flight Avoidance Systems.
>They will see you before you see them.
>
>Mike $$$

Can you describe the In-flight Avoidance System you mention? Does it
rely on the operation of a transponder in the conflicting aircraft?
Is it radar based? Visual based? Please provide more information on
this system.

Tony Cox

unread,
Apr 23, 2004, 12:09:19 PM4/23/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:07og80drrpqsja7el...@4ax.com...

>
> Aircraft operation in the NAS by an uncertificated "pilot" would seem
> to violate FARs.

I suppose that depends on how you define "Aircraft"
and "pilot"...

>
> It's my understanding that it takes a team of about 7 to operate a
> UAV. Perhaps that level of redundancy might mitigate the concerns you
> raise. However, 7 border patrol officers on the ground might be more
> effective in preventing illegal entries.

Now I don't understand the logic. What does a UAV provide
that a 182 doesn't? Is it significantly cheaper to keep in the air?
Do the "team of 7" work for less money than a pilot and a
spotter? Now that's scary....

I do understand the use of UAV in hazardous areas, where there
is enemy fire and/or risk of a pilot being captured. But why go to
all the extra trouble just to police the border?


Ace Pilot

unread,
Apr 23, 2004, 1:30:24 PM4/23/04
to
How is see-and-avoid handled with unmanned weather balloons? Are they
only released in restricted airspace? Seems to me that there are some
parallels with UAVs.

Peter Gottlieb

unread,
Apr 23, 2004, 4:39:43 PM4/23/04
to

"Tony Cox" <t...@coxrt.com> wrote in message
news:Pcbic.8669$e4....@newsread2.news.pas.earthlink.net...

> Now I don't understand the logic. What does a UAV provide
> that a 182 doesn't? Is it significantly cheaper to keep in the air?
> Do the "team of 7" work for less money than a pilot and a
> spotter? Now that's scary....

Surely you can understand how a UAV and support team are much more efficient
at going through taxpayer money than a 172.

Perhaps it is another pork barrel project, or some company has friends in
high places. Because, if they put out the ACTUAL job of searching for
illegal crossings I would find it very hard to believe that it couldn't be
done by a small fleet of properly equipped 172's or 182's and a reasonable
staff of pilots and observers.

Heck, why don't they try giving the job to CAP and see how well that concept
works?


Bob Noel

unread,
Apr 23, 2004, 4:57:04 PM4/23/04
to
In article <Pcbic.8669$e4....@newsread2.news.pas.earthlink.net>, "Tony
Cox" <t...@coxrt.com> wrote:

> Now I don't understand the logic. What does a UAV provide
> that a 182 doesn't?

loiter time.

altitude.

And all potential 182 buyers will appreciate the military
NOT buying up good 182's.

--
Bob Noel

Mike Money

unread,
Apr 23, 2004, 8:40:48 PM4/23/04
to
Larry,

The Predator is equipped with a modified Honeywell ETCAS TPA-81A. The
system responds to Mode 1, 2, 3, 4, A, C, and S. Forward surveillance
has been extended to 360 degrees.

In addition, Predator is data-linked to airborne and ground commands for
control and observation.

Predator is piloted by a ground controller who is assisted by up to six
(6) mission specialist. Each specialists is responsible for the
sensor/system he/she is operating to complete the mission (optical, IR,
armament, etc.). The pilot ground controller is dedicated to flying the
airplane. Some controllers are certified pilots and all controllers
have spent many hours in a simulator.

There are more eyeballs on a Predator and its proximity to everything
than any GA aircraft.

Mike $$$ (PA28)

John T

unread,
Apr 24, 2004, 1:38:37 AM4/24/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:sc8i80lvrfebecamo...@4ax.com

>
>> Frankly, I'd give much better odds to having an in-flight fire or
>> engine failure than a MAC with a remotely piloted aircraft.
>
> How did you arrive at that point of view. Do you have any data to
> support it?

No, I don't. Those are *my* odds I'm offering. :)

> Giving odds or taking chances is an inappropriate approach to air
> safety.

BS. You take chances stepping into the shower. You take chances crossing
the street. You take chances driving to the airport. You take chances
leaving the ground in an aircraft. It's what you do to minimize those
chances that counts and nothing I've seen in your articles re: UAV's leads
me to believe that they're necessarily a significant safety issue. Once you
show me *evidence* of lackadaisical attention to safety by the owners and
operators of those very expensive bits of hardware, then I'll join your
rally. Until then, this is my last post on the issue. It's already gotten
far more attention than it deserves at this stage.

> The "Big Sky" is a total myth. Any rational system that relies upon
> chance to insure air safety is doomed to failure. I hope you're not
> an FAA employee.

hmm... For the record, no, I'm not an FAA employee. However, the "myth" of
the big sky is shattered everytime I go up VFR. For all the VFR flight I've
done, the only time I have ever gotten close to another craft
unintentionally was near an airport. See and avoid? Perhaps, but I don't
recall ever maneuvering to avoid another aircraft during VFR cruise. Also
for the record, you inferred a reliance on chance for safety. I implied no
such thing. Until proven otherwise, I will stand by my assertion that there
are far fewer airplanes in operation (i.e., "Big Sky") in the border areas
under consideration for UAV use, though.

> You questioned my use of the term 'reduced vision standards'.

Yes, I did.

> That
> lead me to believe that you felt that UAV operators would be held to
> the same (not reduced) vision standards as certificated airmen.

I can't help that.

> If
> your questioning of my use of the term 'reduced vision standards' did
> not imply your belief that they UAV operators would be held to the
> same standards as certificated airmen, what were you implying?

I was implying that you have no idea what are the capabilities of these
UAV's you're trying to get us all stirred up about. Nothing more.

> The military has not disclosed to me all the capabilities of their
> UAVs. :-) However, unless there is high-resolutin, color, binocular
> vision in all quadrants, the UAV operators visual capability to see
> and avoid will be substandard to that required of a certificated
> airman.

That may be, but there are ways to compensate. Again, you haven't
demonstrated that the proposed operation of these UAV's will significantly
degrade aviation safety. Come back when you have something more solid than
"omigod they're putting unmanned aircraft in the skies!"

> Your apparent lack of concern for air safety and reliance on chance
> (Big Sky)for aircraft separation betrays your shallow understanding of
> the issue.

It's interesting that you think I have any less concern for aviation safety
than anybody else - much less rely on chance for separation. Larry, you're
demonstrating a serious ignorance here.

> So you feel that a _biased_ (as opposed to _impartial_) governmental
> investigative organization does not warrant reform? Comon' man,
> think!

Now you're trying to change the subject. If you want to talk about
revamping the NTSB, start another thread. This one's dead.

Cub Driver

unread,
Apr 24, 2004, 5:46:37 AM4/24/04
to

>> Do the "team of 7" work for less money than a pilot and a
>> spotter?

You are assuming that the two-man crew of the Cessna is supported by
no one?

This is the government! That can't be right!

all the best -- Dan Ford
email: war...@mailblocks.com (put Cubdriver in subject line)

The Warbird's Forum www.warbirdforum.com
The Piper Cub Forum www.pipercubforum.com
Viva Bush! blog www.vivabush.org

Larry Dighera

unread,
Apr 24, 2004, 10:38:08 AM4/24/04
to
On Sat, 24 Apr 2004 05:38:37 GMT, "John T" <j...@his.isp> wrote in
Message-Id: <06835121dca2a37b...@news.bubbanews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:sc8i80lvrfebecamo...@4ax.com

>> Giving odds or taking chances is an inappropriate approach to air


>> safety.
>
>BS. You take chances stepping into the shower. You take chances crossing
>the street. You take chances driving to the airport. You take chances
>leaving the ground in an aircraft. It's what you do to minimize those
>chances that counts

Fortunately, the chances you cite are not criteria for NAS design.

In engineering a workable NAS I would prefer that the designers employ
methodologies that _insure_ separation of air traffic, not merely
reduce the _chances_ of a MAC. Anything less is irresponsible
negligence.

If reliance on the Big Sky theory were adequate for separating
aircraft, we wouldn't need ATC.

>and nothing I've seen in your articles re: UAV's leads
>me to believe that they're necessarily a significant safety issue. Once you
>show me *evidence* of lackadaisical attention to safety by the owners and
>operators of those very expensive bits of hardware, then I'll join your
>rally.

I'm happy to have you aboard. :-) Here is the information you
request:

http://www.aetc.randolph.af.mil/se2/torch/back/2003/0305/runway.htm
GROUND CREW’S INATTENTION
LEADS TO UNMANNED AIRCRAFT CRASH

LANGLEY AIR FORCE BASE, Va. (ACCNS) — Investigators determined
that pilot error caused an Air Force RQ-1 Predator aircraft to
crash Oct. 25, nine miles west of Indian Springs Air Force
Auxiliary Field, Nev.

The Predator, an unmanned reconnaissance aircraft, was destroyed
upon impact. The loss is estimated at $3.3 million. No one was
injured. The aircraft was assigned to the 11th Reconnaissance
Squadron at Nellis Air Force Base, Nev.

According to an Air Combat Command accident investigation report
released last month, the primary cause of the accident was the
ground crew’s inattention to the aircraft’s altitude.

While trying to enter the Indian Springs flight pattern, the
aircraft was flown over mountainous terrain, obstructing the
datalink and causing the ground crew to lose electronic contact
with the aircraft.

Following failed attempts to regain the link, the pilot executed
emergency procedures designed to safeguard the aircraft; however,
the aircraft impacted mountainous terrain 16 seconds later.


http://www.af.mil/news/Feb2001/n20010202_0157.shtmlOfficials

02/02/01
Officials release RQ-1L Predator RQ-1L Predator accident report
the accident resulted from operator error.

the pilot -- who flies the aircraft from a ground control station
-- inadvertently cleared the primary control module's random
access memory. As a result, the Predator lost its data link
connection with the ground control station.


http://www.globalsecurity.org/military/library/news/2003/02/mil-030219-acc01.htm
releases RQ-1 accident report

In-Depth Coverage

Released: Feb. 19, 2003

LANGLEY AIR FORCE BASE, Va. (ACCNS) -- Air Force investigators
have determined that human error caused an RQ-1 Predator aircraft
to crash Sept. 17 at a classified forward-operating location in
Southwest Asia.

The Predator, which is an unmanned reconnaissance aircraft, was
destroyed upon impact. The loss is estimated at $3.2 million. No
one was injured in the accident. The aircraft was assigned to the
11th Reconnaissance Squadron at Nellis Air Force Base, Nev.

According to an Air Combat Command accident investigation report
released today, the primary cause of the accident was that the
pilot unintentionally flew the aircraft into a hazardous cloud.

The pilot lost communication with the aircraft several times, but
was able to re-establish communication twice. However, the
aircraft failed to respond to the pilot’s commands, indicating the
flight control computers were disabled by the hazardous weather
conditions


http://www.globalsecurity.org/intell/systems/predator.htm
As of 31 October 2001 the Air Force had received a total of 68 air
vehicles, and had lost 19 due to mishaps or losses over enemy
territory, including four over enemy territory in Kosovo. A good
number of them were lost due to operator error, since it is hard
to land the UAV. The operator has the camera pointing out the
front of the plane, but he really has lost a lot of situational
awareness that a normal pilot would have of where the ground is
and where the attitude of his aircraft is.

The CIA has a small number of the armed drones. Newer versions of
the Predator, at $4.5 million each, are being produced at a rate
of about two aircraft a month.


http://www.reviewjournal.com/lvrj_home/2002/Oct-31-Thu-2002/news/19962521.html
Thursday, October 31, 2002
Las Vegas Review-Journal

May 17 crash of unmanned spy plane blamed on human error

Investigators have blamed the May 17 crash of an unmanned Predator
spy plane in Southwest Asia on human error, saying one of the
plane's tail control mechanisms had been improperly assembled by
the manufacturer, according to an Air Force statement Wednesday.

The remote-controlled RQ-1 Predator was assigned to Nellis Air
Force Base's 15th Reconnaissance Squadron in Indian Springs.

The plane, which had been deployed as part of the 386th
Expeditionary Group, went down "near a classified forward
operating location" in Southwest Asia, said the statement from Air
Combat Command headquarters at Langley Air Force Base, Va.

The loss of the plane made by General Atomics of San Diego is
valued at $3.3 million, the statement said.

Air Force investigators determined that incorrect assembly of the
"right tail plane control servo" was the sole cause of the
accident, the statement said.

A spokesman for Air Combat Command said Air Force officials are
still probing Friday's crash of a Predator during a training
mission near Indian Springs. That plane was assigned to Nellis'
11th Reconnaissance Squadron.

http://www2.acc.af.mil/accnews/aug01/01267.html
Released: Aug. 16, 2001

RQ-1 Predator accident report released

The RQ-1 Predator is a medium-altitude, long-endurance unmanned
aerial vehicle system. The Predator is a system, not just an
aircraft. The fully operational system consists of four air
vehicles (with sensors), a ground control station, a Predator
primary satellite link communication suite and 55 people.
(Courtesy photo)

LANGLEY AIR FORCE BASE, Va. (ACCNS) -- Officials investigating the
March 30 crash of an RQ-1L Predator unmanned aerial vehicle have
determined the accident resulted from operator error.

According to the Accident Investigation Board report released
today by Air Combat Command, the Predator experienced an icing
problem and the pilot was unable to maintain control of the
aircraft.

The Predator, which belonged to the 11th Reconnaissance Squadron
at Nellis AFB, Nev., was supporting the Kosovo Stabilization
Force. There were no injuries or fatalities. The Predator was
destroyed upon impact.

According to the report, the pilot recognized the icing problem,
but failed to immediately execute critical checklist steps for
pitot static system failure. The pitot static system uses air and
static pressure to determine the aircraft’s altitude and airspeed.
There is also substantial evidence that nonuse of the pitot static
heating system was a substantially contributing factor in this
mishap.

http://www2.acc.af.mil/accnews/dec99/990383.html
Released: December 23, 1999

RQ-1 Predator accident report released

LANGLEY AIR FORCE BASE, VA. (ACCNS) -- Officials investigating the
April 18 crash of an RQ-1 Predator unmanned aerial vehicle near
Tuzla Air Base, Bosnia, have determined the accident resulted from
a combination of mechanical and human factors.

The Predator, which belonged to the 11th Reconnaissance Squadron
at Nellis Air Force Base, Nev., was returning from a
reconnaissance mission over Kosovo in support of Operation Allied
Force. It was destroyed upon impact.

According to the Accident Investigation Board report released Dec.
22 by Air Combat Command, the Predator experienced a fuel problem
during its descent into Tuzla. Upon entering instrument
meteorological conditions and experiencing aircraft icing, the
Predator lost engine power.

The two Predator pilots, who control the aircraft from a ground
station, executed critical action procedures but were unable to
land the aircraft safely. It crashed in a wooded area four miles
south of Tuzla AB.

According to the report, the pilots' attention became too focused
on flying the Predator in icing and weather conditions they had
rarely encountered. The report also cites lack of communication
between the two pilots during the flight emergency as a cause of
the accident.

For more information, please contact the Air Combat Command Public
Affairs office at (757) 764-5994 or e-mail acc...@langley.af.mil.

http://www2.acc.af.mil/accnews/apr01/01127.html
Released: April 13, 2001

Predator accident report released

LANGLEY AIR FORCE BASE, Va. (ACCNS) -- An RQ-1K Predator unmanned
aerial vehicle crashed Oct. 23 in Kosovo as a result of mechanical
failure, according to accident investigators.

The Predator is an unmanned reconnaissance aircraft used to survey
battlefields and return video footage and radar data. The accident
happened about 180 miles southeast of Tuzla Air Base, Bosnia,
where the aircraft was based. The Predator was part of an
Operation Joint Forge reconnaissance mission over Kosovo and was
assigned to the 11th Reconnaissance Squadron, Nellis Air Force
Base, Nev.

According to Air Combat Command's Accident Investigation Board
report released Thursday, the accident resulted from mechanical
failure in the UAV's propeller control system. Investigators found
substantial evidence indicating errors during maintenance on the
propeller control system on Sept. 28 played a critical role in the
accident. Evidence showed that certain components of the propeller
assembly were not adequately lubricated; in addition, a key bolt
was stripped and had not been tightened properly. These errors
likely led to the accident, according to the lead investigator.

>Until then, this is my last post on the issue.

I suppose that means you'll be continuing to post to this thread. :-)

>It's already gotten far more attention than it deserves at this stage.

At what stage do you feel public scrutiny of UAV operation in civil
airspace would be appropriate? Oh I forgot. You want to see NTSB
reports before you consider the hazard posed by UAV operation in civil
airspace. Brilliant! :-)

>> The "Big Sky" is a total myth. Any rational system that relies upon
>> chance to insure air safety is doomed to failure. I hope you're not
>> an FAA employee.
>
>hmm... For the record, no, I'm not an FAA employee. However, the "myth" of
>the big sky is shattered everytime I go up VFR. For all the VFR flight I've
>done, the only time I have ever gotten close to another craft
>unintentionally was near an airport.

There are several airports very near the US/Mexico boarder.

>See and avoid? Perhaps, but I don't
>recall ever maneuvering to avoid another aircraft during VFR cruise.

I suppose encountering conflicting air traffic is more likely in
congested airspace, however I've often had traffic in close proximity
over the Mojave Desert. The sky is getting smaller all the time as
the military grabs more and airline traffic increases require
increasing the size of class, B, C, & D areas.

>Also for the record, you inferred a reliance on chance for safety. I implied no
>such thing.

Citing the Big Sky theory as your separation methodology of choice for
UAV operation seems to contradict your denial of reliance on chance
for air safety.

>Until proven otherwise, I will stand by my assertion that there
>are far fewer airplanes in operation (i.e., "Big Sky") in the border areas
>under consideration for UAV use, though.

How does the number of aircraft operating in a given area justify
chance as the chosen method of separating them?

>> You questioned my use of the term 'reduced vision standards'.
>
>Yes, I did.
>
>> That
>> lead me to believe that you felt that UAV operators would be held to
>> the same (not reduced) vision standards as certificated airmen.
>
>I can't help that.
>
>> If
>> your questioning of my use of the term 'reduced vision standards' did
>> not imply your belief that they UAV operators would be held to the
>> same standards as certificated airmen, what were you implying?
>
>I was implying that you have no idea what are the capabilities of these
>UAV's you're trying to get us all stirred up about. Nothing more.

No idea? They are unmanned. I believe that a pilot is certified to
meet vision standards that are impossible to meet with synthetic
vision.

>> The military has not disclosed to me all the capabilities of their
>> UAVs. :-) However, unless there is high-resolutin, color, binocular
>> vision in all quadrants, the UAV operators' visual capability to see
>> and avoid will be substandard to that required of a certificated
>> airman.
>
>That may be, but there are ways to compensate.

Please don't withhold your description of those "ways to compensate."
I am most interested to know to which 'ways' you allude.

>Again, you haven't
>demonstrated that the proposed operation of these UAV's will significantly
>degrade aviation safety. Come back when you have something more solid than
>"omigod they're putting unmanned aircraft in the skies!"

See the citations of numerous UAV operator error crashes I provided
above. These mishaps enumerate operator inattention, improper
operator commands, loss of control due to data link failure as a
result of flying into a cloud, operator loss of situational awareness,
operator failure to recognize pitot static system failure, incorrect
assembly of control servo, operator lack of experience in IMC, lack of
lubrication and improper assembly...

>> Your apparent lack of concern for air safety and reliance on chance
>> (Big Sky)for aircraft separation betrays your shallow understanding of
>> the issue.
>
>It's interesting that you think I have any less concern for aviation safety
>than anybody else - much less rely on chance for separation. Larry, you're
>demonstrating a serious ignorance here.

I'm just reading what you wrote. If you meant something else, you
should have said something else.

>> So you feel that a _biased_ (as opposed to _impartial_) governmental
>> investigative organization does not warrant reform? Comon' man,
>> think!
>
>Now you're trying to change the subject. If you want to talk about
>revamping the NTSB, start another thread. This one's dead.

It was you who first mentioned the NTSB not me:

Message-ID: <36b0f2ceda472b1f...@news.bubbanews.com>


As for your subject line question, I'd wait for an NTSB ruling
before passing judgment on that.

But I suppose you forgot what you said a day and a half ago.


Larry Dighera

unread,
Apr 24, 2004, 11:06:41 AM4/24/04
to
On Fri, 23 Apr 2004 19:40:48 -0500, mmw...@webtv.net (Mike Money)
wrote in Message-Id: <1731-408...@storefull-3317.bay.webtv.net>:

>Larry,
>
>The Predator is equipped with a modified Honeywell ETCAS TPA-81A. The
>system responds to Mode 1, 2, 3, 4, A, C, and S. Forward surveillance
>has been extended to 360 degrees.

http://www.honeywelltcas.com/etcas_tpa81a.htm
System Operation
ETCAS provides two modes of operation. The basic mode is ACAS II
which is the same as TCAS II with Change 7.0 software and is RVSM
compatible. In addition to the standard TCAS functions of
situational awareness, traffic alert and resolution advisories,
the Honeywell ETCAS provides a formation mode. This formation mode
allows aircraft operators to locate, identify, rendezvous with and
maintain flight formation with aircraft equipped with a variety of
identification systems, including Identification Friend and
Fo(IFF), Modes 1, 2, 3, and 4, Mode A, Mode C and Modes S
transponder equipped private, commercial and military aircraft.

In order for UAV operators to rely upon the Honeywell ETCAS for
aircraft separation, _all_ aircraft would have to be transponder
equipped, and FARs would have to be changed to mandate transponder use
at all times while airborne. So while TCAS is definitely part of the
solution to aircraft separation, it would not separate UAVs from
aircraft without electrical systems, nor those operating in airspace
where transponder operation is not mandated by regulation.

>In addition, Predator is data-linked to airborne and ground commands for
>control and observation.

I'd like to know more about that.

>Predator is piloted by a ground controller who is assisted by up to six
>(6) mission specialist. Each specialists is responsible for the
>sensor/system he/she is operating to complete the mission (optical, IR,
>armament, etc.). The pilot ground controller is dedicated to flying the
>airplane. Some controllers are certified pilots and all controllers
>have spent many hours in a simulator.

How will the flying public feel about sharing the sky with
uncertificated UAV operators with lots of sim time? Shall we now
permit gamers with lots of MS Flight Simulator time to ply the
nation's skies? Yikes!

>There are more eyeballs on a Predator and its proximity to everything
>than any GA aircraft.
>
>Mike $$$ (PA28)

What is the aggregate cost for all those eyeballs?

What is the cost of two man Cessna 182 patrol?

Are there any eyeballs aboard the UAV that meet the vision
requirements of a certificated airman: 20/20 binocular color vision?

Before the government starts operating UAVs among the flying public,
they need to insure UAVs will meet the same or better criteria they
currently demand of airman. Anything less is criminal negligence.


Peter Gottlieb

unread,
Apr 24, 2004, 11:12:29 AM4/24/04
to

"John T" <j...@his.isp> wrote in message
news:06835121dca2a37b...@news.bubbanews.com...

>
> However, the "myth" of
> the big sky is shattered everytime I go up VFR. For all the VFR flight
I've
> done, the only time I have ever gotten close to another craft
> unintentionally was near an airport. See and avoid? Perhaps, but I don't
> recall ever maneuvering to avoid another aircraft during VFR cruise.

Fly up here in the NY Metro area for a few years and see how you feel then.
Around here you are *always* close to another airport.


Peter Gottlieb

unread,
Apr 24, 2004, 11:13:41 AM4/24/04
to

"Cub Driver" <war...@mailblocks.com> wrote in message
news:ppdk801skm7n4t2ls...@4ax.com...

>
> >> Do the "team of 7" work for less money than a pilot and a
> >> spotter?
>
> You are assuming that the two-man crew of the Cessna is supported by
> no one?
>
> This is the government! That can't be right!
>

Oh yeah. For a moment I foolishly thought it would be run as a business.


Larry Dighera

unread,
Apr 24, 2004, 11:15:39 AM4/24/04
to
On Fri, 23 Apr 2004 16:09:19 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <Pcbic.8669$e4....@newsread2.news.pas.earthlink.net>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:07og80drrpqsja7el...@4ax.com...
>>
>> Aircraft operation in the NAS by an uncertificated "pilot" would seem
>> to violate FARs.
>
>I suppose that depends on how you define "Aircraft"
>and "pilot"...
>

Ummm..

Pilot: A person who holds the appropriate category, class, and type
rating, if appropriate, for the conduct of the flight.

Aircraft: A device that is used or intended to be used for flight in
the air.

>>
>> It's my understanding that it takes a team of about 7 to operate a
>> UAV. Perhaps that level of redundancy might mitigate the concerns you
>> raise. However, 7 border patrol officers on the ground might be more
>> effective in preventing illegal entries.
>
>Now I don't understand the logic. What does a UAV provide
>that a 182 doesn't? Is it significantly cheaper to keep in the air?
>Do the "team of 7" work for less money than a pilot and a
>spotter? Now that's scary....

Exactly. There have to be undisclosed reasons for deploying UAVs.

>I do understand the use of UAV in hazardous areas, where there
>is enemy fire and/or risk of a pilot being captured. But why go to
>all the extra trouble just to police the border?

Perhaps the DHS is using the UAV for border patrol duty scenario as a
more publicly acceptable vehicle to introduce UAV surveillance nation
wide, because UAV use doesn't seem to make economic nor safety sense
for domestic peacetime operation.

Tony Cox

unread,
Apr 24, 2004, 11:42:46 AM4/24/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:s2ii80lmos14j0c3p...@4ax.com...

> On Fri, 23 Apr 2004 16:09:19 GMT, "Tony Cox" <t...@coxrt.com> wrote in
> Message-Id: <Pcbic.8669$e4....@newsread2.news.pas.earthlink.net>:
> >
> >I suppose that depends on how you define "Aircraft"
> >and "pilot"...
> >
>
> Ummm..
>
> Pilot: A person who holds the appropriate category, class, and type
> rating, if appropriate, for the conduct of the flight.
>
> Aircraft: A device that is used or intended to be used for flight in
> the air.

I was being "Clintonesque". These definitions from the FAR's? I
couldn't find them on a quick scan.


Barry

unread,
Apr 24, 2004, 11:44:10 AM4/24/04
to
> Fortunately, the chances you cite are not criteria for NAS design.
>
> In engineering a workable NAS I would prefer that the designers employ
> methodologies that _insure_ separation of air traffic, not merely
> reduce the _chances_ of a MAC. Anything less is irresponsible
> negligence.

In any system, there's always a small probability that a catastrophe will
occur. Aircraft certification rules and separation standards acknowledge this
and are established to keep the risk acceptably low. For example, for lateral
separation of two aircraft traveling at the same flight level on parallel
routes, the Target Level of Safety (TLS) set by ICAO (with FAA participation)
is 5 x 10^-9 per flight hour. That is, loss of lateral separation should
lead to no more than one accident every 200 million flight hours. The TLS is
not zero. Some people don't like to accept this, but it's just not realistic
to insist on zero risk.

Barry

Larry Dighera

unread,
Apr 24, 2004, 11:53:49 AM4/24/04
to
On Sat, 24 Apr 2004 15:42:46 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <WVvic.7462$eZ5....@newsread1.news.pas.earthlink.net>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:s2ii80lmos14j0c3p...@4ax.com...
>> On Fri, 23 Apr 2004 16:09:19 GMT, "Tony Cox" <t...@coxrt.com> wrote in
>> Message-Id: <Pcbic.8669$e4....@newsread2.news.pas.earthlink.net>:
>> >
>> >I suppose that depends on how you define "Aircraft"
>> >and "pilot"...
>> >
>>
>> Ummm..
>>
>> Pilot: A person who holds the appropriate category, class, and type
>> rating, if appropriate, for the conduct of the flight.
>>
>> Aircraft: A device that is used or intended to be used for flight in
>> the air.
>
>I was being "Clintonesque".

I know, but today that's the way government works.

>These definitions from the FAR's? I
>couldn't find them on a quick scan.
>

FAR Part I


Larry Dighera

unread,
Apr 24, 2004, 11:57:19 AM4/24/04
to
On Sat, 24 Apr 2004 11:44:10 -0400, "Barry" <n...@body.com> wrote in
Message-Id: <abadnS-0nOP...@comcast.com>:

Thank you for the information.

How would the TLS be affected if the Big Sky theory were relied upon
for aircraft separation as John T. suggested?

Tony Cox

unread,
Apr 24, 2004, 12:09:26 PM4/24/04
to
"Mike Money" <mmw...@webtv.net> wrote in message
news:1731-408...@storefull-3317.bay.webtv.net...

>
> The Predator is equipped with a modified Honeywell ETCAS TPA-81A. The
> system responds to Mode 1, 2, 3, 4, A, C, and S. Forward surveillance
> has been extended to 360 degrees.

What do you mean by 360 degrees? What is the resolution? Better or
worse than someone with 20/40 vision? Is there collision detection
software analyzing the incoming video, or does it just rely on the ground
based operator to see what's going on.

And of course not all GA aircraft are equipped with TCAS, nor are
they required to be.

>
> In addition, Predator is data-linked to airborne and ground commands for
> control and observation.

From the crash reports that Larry provided, this seems to be a ground
link which is easily obscured by terrain. I'd have thought some satellite
link would be better.

>
> Predator is piloted by a ground controller who is assisted by up to six
> (6) mission specialist. Each specialists is responsible for the
> sensor/system he/she is operating to complete the mission (optical, IR,
> armament, etc.). The pilot ground controller is dedicated to flying the
> airplane. Some controllers are certified pilots and all controllers
> have spent many hours in a simulator.

"Some" are certified pilots??? Come on now.

I bet there are millions of little weenies with hundreds of hours
of Microsoft FS under their belts, but I certainly wouldn't want them
flying around in the same sky as me and my passengers.

>
> There are more eyeballs on a Predator and its proximity to everything
> than any GA aircraft.

It's not "eyeballs on a Predator" that concern me. It's the
eyeballs the Predator has looking out for other traffic and
the competence of those interpreting what they see which
is the safety concern.

In the final analysis, the operator of a Predator just has his
job on the line; I have my life on the line, and that of my
passengers.

Tell you what. How about fitting operators with a helmet
that has a built-in gun pointing directly into his head? If they
hit another plane, the gun goes off.

Tony Cox

unread,
Apr 24, 2004, 12:19:31 PM4/24/04
to
"Ace Pilot" <acepi...@mypersonalemail.com> wrote in message
news:c0e5a2d8.04042...@posting.google.com...

> How is see-and-avoid handled with unmanned weather balloons? Are they
> only released in restricted airspace? Seems to me that there are some
> parallels with UAVs.

Aircraft are obliged to give priority to balloons, unmanned or not.
No one has asserted that UAVs have priority over aircraft.

Anyway, balloons don't converge on you from your blind side.


Tony Cox

unread,
Apr 24, 2004, 12:20:37 PM4/24/04
to
"Peter Gottlieb" <ng-repl...@icq.net> wrote in message
news:jafic.280$GN3.2...@news4.srv.hcvlny.cv.net...

>
> Heck, why don't they try giving the job to CAP and see how well that
concept
> works?

Around here (Las Vegas, NV), they do. Not sure how successful
they are...


Barry

unread,
Apr 24, 2004, 12:33:16 PM4/24/04
to

> How would the TLS be affected if the Big Sky theory were relied upon
> for aircraft separation as John T. suggested?

I'm not very familiar with this subject, but you can read "Safety
Considerations for Operation of Small Unmanned Aerial Vehicles in
Civil Airspace" produced by the MIT International Center for Air
Transportation:

http://icat-server.mit.edu/Library/fullRecord.cgi?idDoc=205

They studied both midair collisions and exposure to people on the ground. The
relevant conclusions for midairs:

Significant Amount of Airspace with Exposure Risk
Below the Target Level of Safety
- Areas around major airports are above the TLS

Opportunities may exist to allow a class of small
UAV’s to operate with limited restrictions
- Limiting operation in airspace near airports
may achieve TLS

Mitigation Strategies Are Available to Further Reduce the Risk
-Vehicles can be designed with capabilities to limit
likelihood of midair collisions

Barry

Tony Cox

unread,
Apr 24, 2004, 12:41:33 PM4/24/04
to
"John T" <j...@his.isp> wrote in message
news:06835121dca2a37b...@news.bubbanews.com...

>
> For all the VFR flight I've
> done, the only time I have ever gotten close to another craft
> unintentionally was near an airport. See and avoid? Perhaps, but I don't
> recall ever maneuvering to avoid another aircraft during VFR cruise.

Perhaps your good fortune or lack of attention has lulled you into
a false sense of security. In my 500+ hours, I've been almost dinged
twice -- once some 20 miles out from Pasa Rubles and once in the
middle of nowhere. Both were near head-ons. And I'm willing to bet
that there have been more I've not been aware of, since in both cases
the occupants displayed no reaction to my presence whatsoever.


Stan Gosnell

unread,
Apr 24, 2004, 1:10:42 PM4/24/04
to
"Tony Cox" <t...@coxrt.com> wrote in
news:1Nwic.7528$eZ5....@newsread1.news.pas.earthlink.net:

> Perhaps your good fortune or lack of attention has lulled you into
> a false sense of security. In my 500+ hours, I've been almost dinged
> twice -- once some 20 miles out from Pasa Rubles and once in the
> middle of nowhere. Both were near head-ons. And I'm willing to bet
> that there have been more I've not been aware of, since in both cases
> the occupants displayed no reaction to my presence whatsoever.

I've had dozens of close calls, several of which required very abrupt
maneuvers to avoid collision, most of them far from airports. The big sky
theory is just that, a theory.

--
Regards,

Stan

Tony Cox

unread,
Apr 24, 2004, 4:08:29 PM4/24/04
to
"Barry" <n...@body.com> wrote in message
news:5tadnXmsPdS...@comcast.com...

This study attempts to 'bound' the danger through a Bayesian
analysis of engine failure probability and chances of hitting
something at random in the airspace 'per flight hour'. In high
traffic areas, the probability is small (10-8). But the total
accident rate will depend on how many of these things are
flying around. There is nothing about 'accountability' in the
"Mitigation Strategies", which is very odd since accountability
looms very large in current aviation practice (and FAA regulation).

I'm concerned that the model for this sees a UAV "pilot" as a
sort of hands-on air traffic controller, rather than as a proper
"pilot" with the attendant certification and responsibility
requirements. This is a major departure from existing practice,
and potentially devastating for GA.


William W. Plummer

unread,
Apr 25, 2004, 4:34:31 PM4/25/04
to

"Tony Cox" <t...@coxrt.com> wrote in message
news:1Pzic.7869$eZ5....@newsread1.news.pas.earthlink.net...

I had an opportunity to speak with a Marine who operates UAVs as the remote
pilot. He said he and others doing that job must be instrument rated pilots
and the UAV must be on an IFR flight plan. Given that, why would the
accident rate for UAVs be any different than normal IFR traffic?


Tony Cox

unread,
Apr 26, 2004, 7:58:48 AM4/26/04
to
"William W. Plummer" <William.Plu...@alum.mit.edu> wrote in message
news:rhVic.20812$YP5.1530448@attbi_s02...

>
> I had an opportunity to speak with a Marine who operates UAVs as the
remote
> pilot. He said he and others doing that job must be instrument rated
pilots
> and the UAV must be on an IFR flight plan.

That may be true in his case (instrument rated pilot), but it isn't
required according to Larry's original post.

> Given that, why would the
> accident rate for UAVs be any different than normal IFR traffic?

1) Conventional traffic must "See and avoid" when in VMC even if
flying IFR.

2) The remote "pilot" doesn't need to keep alert to the extent that
the rest of us do because his life isn't on the line.

Steven P. McNicoll

unread,
Apr 26, 2004, 4:50:12 PM4/26/04
to

"Tony Cox" <t...@coxrt.com> wrote in message
news:NRThc.7106$e4....@newsread2.news.pas.earthlink.net...
>
> It is if the pilot can't scan for traffic or search for a
> suitable emergency landing site.
>

Is the UAV pilot not able to scan for traffic or search for a suitable
emergency landing site?


>
> In any case, an aircraft "in distress" is only allowed to violate the
> FAR's as far as necessary to deal with the emergency. I'd have a
> hard time proving reasonable violation of "See and
> Avoid" in the simple case of engine failure.
>

We're not talking about violating any FARs.


>
> You're proposing that these UAV's
> can simply ignore this rule because they're supposed to be in class
> A all the time.
>

I proposed nothing at all like that.


Tony Cox

unread,
Apr 26, 2004, 6:18:09 PM4/26/04
to
"Steven P. McNicoll" <ronca...@nospamearthlink.net> wrote in message
news:8Cejc.3612$g31...@newsread2.news.atl.earthlink.net...

>
> "Tony Cox" <t...@coxrt.com> wrote in message
> news:NRThc.7106$e4....@newsread2.news.pas.earthlink.net...
> >
> > It is if the pilot can't scan for traffic or search for a
> > suitable emergency landing site.
> >
>
> Is the UAV pilot not able to scan for traffic or search for a suitable
> emergency landing site?

Unless the entire system (camera=>ground station=>operator)
can match private pilot vision requirements, and is as flexible and
responsive as a qualified private pilot, then the UAV will always
be inherently less safe.

The MIT report does its statistical analysis assuming that the UAV
will just blunder into whatever happens to be in its airspace, and that
if the engine fails it'll come down like a WW2 doodlebug. As reported
before, "collision avoidance" seems to boil down to dodging aircraft
with active transponders. It doesn't appear from anything said so
far in the thread that anyone in the UAV program has considered the
situation from a VFR pilot's point of view.

I have no problem with UAVs -- if their use is confined to
MOAs or MTRs or anywhere else where we have a 'heads up'
that some 'unusual' traffic can be expected. But if these things are
buzzing around in the NAS, then it's reasonable that they obey the
same rules as the rest of us. Otherwise, lets just dispense with the
PP vision & medical requirements, junk 91.113(b) and have
a free-for-all.

> >
> > You're proposing that these UAV's
> > can simply ignore this rule because they're supposed to be in class
> > A all the time.
> >
>
> I proposed nothing at all like that.

I assumed this was what you meant when you said
in a response to Larry:- "There is no altitude
reference outside of Class A airspace, so presumably
see-and-avoid is not an issue.". Did I misunderstand you?

Larry Dighera

unread,
Apr 26, 2004, 11:42:33 PM4/26/04
to
On Mon, 26 Apr 2004 20:50:12 GMT, "Steven P. McNicoll"
<ronca...@nospamearthlink.net> wrote in Message-Id:
<8Cejc.3612$g31...@newsread2.news.atl.earthlink.net>:

>
>"Tony Cox" <t...@coxrt.com> wrote in message
>news:NRThc.7106$e4....@newsread2.news.pas.earthlink.net...
>>
>> It is if the pilot can't scan for traffic or search for a
>> suitable emergency landing site.
>>
>
>Is the UAV pilot not able to scan for traffic or search for a suitable
>emergency landing site?
>

It's doubtful the ground based UAV operators' vision systems would
function equal to or better than that required meet the Part 61
requirements. It would be necessary for the systems to provide color,
binocular, 20/20 vision in all quadrants, in real-time.

Anything less would restrict a lot more airspace or negatively impact
public air safety (or both), wouldn't it?

Larry Dighera

unread,
Apr 26, 2004, 11:51:20 PM4/26/04
to
On Mon, 26 Apr 2004 22:18:09 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <BUfjc.13850$e4....@newsread2.news.pas.earthlink.net>:

>
>I have no problem with UAVs -- if their use is confined to
>MOAs or MTRs or anywhere else where we have a 'heads up'
>that some 'unusual' traffic can be expected. But if these things are
>buzzing around in the NAS, then it's reasonable that they obey the
>same rules as the rest of us. Otherwise, lets just dispense with the
>PP vision & medical requirements, junk 91.113(b) and have
>a free-for-all.

I have a problem with them (as I understand them to currently be
equipped) in all Joint Use airspace, including MOAs and MTRs. To
accommodate UAVs in the NAS, considerably more systems analysis and
testing will need to be conducted before it would even be appropriate
to consider a UAV NPRM. Certainly the FARs would have to be revised
to accommodate UAVs. The MIT project at least provides the means, if
not the best methods, of one getting his arms around the issue.


Bob Noel

unread,
Apr 27, 2004, 6:49:47 AM4/27/04
to
In article <d4lr80h0avuekflsk...@4ax.com>, Larry Dighera
<LDig...@att.net> wrote:

> It's doubtful the ground based UAV operators' vision systems would
> function equal to or better than that required meet the Part 61
> requirements.

UAVs have limited payloads. The UAV mission is not to carry
heavy equipment to scan for other aircraft.

--
Bob Noel

Bob Jones

unread,
Apr 27, 2004, 8:40:11 AM4/27/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:d4lr80h0avuekflsk...@4ax.com

>
> It's doubtful the ground based UAV operators' vision systems would
> function equal to or better than that required meet the Part 61
> requirements. It would be necessary for the systems to provide color,
> binocular, 20/20 vision in all quadrants, in real-time.

All quadrants? What about all the aircraft with no rearward visibility?
How do those pilots scan all quadrants with 20/20 vision in real-time?

By the way, where in Part 61 are the vision requirements specified?


Larry Dighera

unread,
Apr 27, 2004, 9:44:47 AM4/27/04
to
On Tue, 27 Apr 2004 12:40:11 GMT, "Bob Jones" <bjo...@erols.com> wrote
in Message-Id: <906be260063b3f03...@news.teranews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:d4lr80h0avuekflsk...@4ax.com
>>
>> It's doubtful the ground based UAV operators' vision systems would
>> function equal to or better than that required meet the Part 61
>> requirements. It would be necessary for the systems to provide color,
>> binocular, 20/20 vision in all quadrants, in real-time.
>
>All quadrants? What about all the aircraft with no rearward visibility?
>How do those pilots scan all quadrants with 20/20 vision in real-time?

I was taught to scan for traffic to the rear of the side to which I
intend to turn before making the turn.

>By the way, where in Part 61 are the vision requirements specified?

Here:
http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr&sid=e7716ce2eff8b556898f8532c61b11bf&rgn=div8&view=text&node=14:2.0.1.1.5.4.1.2&idno=14

Bob Jones

unread,
Apr 27, 2004, 3:26:56 PM4/27/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:6bos80pgt5n0m7vb3...@4ax.com

>
> I was taught to scan for traffic to the rear of the side to which I
> intend to turn before making the turn.

That may be, but do you perform this scan in all quadrants in real-time? Or
only when you need it? My point is you appear to be placing a higher
standard on UAV's than are placed on existing aircraft/pilots.

>> By the way, where in Part 61 are the vision requirements specified?
>
> Here:

> [really long URL]

Ah, Part *67*, just as I thought. You were either lying or ignorant when
you made this claim:
"It's doubtful the...UAV operators' vision systems would function equal to


or better than that required meet the Part 61 requirements."

As near as I can tell, the words "vision", "eyesight" and "sight" don't even
appear in Part 61. Further, you admitted elsewhere in this thread that you
have no idea what the capabilities of these systems are (yet more
ignorance).

But why let the facts get in the way of your quixotic quest, eh?


Bob Jones

unread,
Apr 27, 2004, 4:11:17 PM4/27/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:r1qk80dlc7gk2silf...@4ax.com
>
>> Once you show me *evidence* of lackadaisical attention to safety by
>> the owners and operators of those very expensive bits of hardware,
>
> I'm happy to have you aboard. :-) Here is the information you
> request:

Interesting that none of those cites indicate anything approaching "evidence
of lackadisical attention to safety".

Lessee... Inadvertent cloud entry. Reference to difficulty in landing.
Faulty assembly. Icing encounters (two of those). Mechanical failure due
to inadequate lubrication.

Sounds like a reading from the NTSB database.

The point is that none of these equate to "lackadaisical attention to safety
by the owners and operators".


Larry Dighera

unread,
Apr 27, 2004, 7:34:59 PM4/27/04
to
On Tue, 27 Apr 2004 20:11:17 GMT, "Bob Jones" <bjo...@erols.com> wrote
in Message-Id: <3604ef10582378aa...@news.teranews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:r1qk80dlc7gk2silf...@4ax.com
>>
>>> Once you show me *evidence* of lackadaisical attention to safety by
>>> the owners and operators of those very expensive bits of hardware,
>>
>> I'm happy to have you aboard. :-) Here is the information you
>> request:
>
>Interesting that none of those cites indicate anything approaching "evidence
>of lackadisical attention to safety".

Obviously our assessments differ.

>Lessee... Inadvertent cloud entry.

http://www.globalsecurity.org/military/library/news/2003/02/mil-030219-acc01.htm
releases RQ-1 accident report
According to an Air Combat Command accident investigation report
released today, the primary cause of the accident was that the
pilot unintentionally flew the aircraft into a hazardous cloud.

FARs prescribe limits on how close to clouds a VFR flight may be, and
a certificated pilot is trained in recovery techniques.


>Reference to difficulty in landing.

http://www2.acc.af.mil/accnews/dec99/990383.html
The two Predator pilots, who control the aircraft from a ground
station, executed critical action procedures but were unable to
land the aircraft safely. It crashed in a wooded area four miles
south of Tuzla AB.

According to the report, the pilots' attention became too focused
on flying the Predator in icing and weather conditions they had
rarely encountered. The report also cites lack of communication
between the two pilots during the flight emergency as a cause of
the accident.


>Faulty assembly.

http://www2.acc.af.mil/accnews/apr01/01127.html
According to Air Combat Command's Accident Investigation Board
report released Thursday, the accident resulted from mechanical
failure in the UAV's propeller control system. Investigators found
substantial evidence indicating errors during maintenance on the
propeller control system on Sept. 28 played a critical role in the
accident. Evidence showed that certain components of the propeller
assembly were not adequately lubricated; in addition, a key bolt
was stripped and had not been tightened properly. These errors
likely led to the accident, according to the lead investigator.

As you will note, the maintenance personnel did not apply the same
diligence in maintaining the unmanned aircraft as is required by
regulation for manned aircraft.


>Icing encounters (two of those).

http://www2.acc.af.mil/accnews/aug01/01267.html
LANGLEY AIR FORCE BASE, Va. (ACCNS) -- Officials investigating the
March 30 crash of an RQ-1L Predator unmanned aerial vehicle have
determined the accident resulted from operator error.

According to the Accident Investigation Board report released
today by Air Combat Command, the Predator experienced an icing
problem and the pilot was unable to maintain control of the
aircraft.

According to the report, the pilot recognized the icing problem,
but failed to immediately execute critical checklist steps for
pitot static system failure.


>Mechanical failure due to inadequate lubrication.

Yes, and a stripped bolt. That sort of poor maintenance practice is
not permitted manned aircraft.


http://www2.acc.af.mil/accnews/apr01/01127.html
According to Air Combat Command's Accident Investigation Board
report released Thursday, the accident resulted from mechanical
failure in the UAV's propeller control system. Investigators found
substantial evidence indicating errors during maintenance on the
propeller control system on Sept. 28 played a critical role in the
accident. Evidence showed that certain components of the propeller
assembly were not adequately lubricated; in addition, a key bolt
was stripped and had not been tightened properly. These errors
likely led to the accident, according to the lead investigator.


>Sounds like a reading from the NTSB database.

No. It sounds like you failed to appreciate the operational errors
committed by the ground based UAV operators and maintenance personnel.

>The point is that none of these equate to "lackadaisical attention to safety
>by the owners and operators".


By omitting the citations below, you publicly demonstrate your biased
thinking:

http://www.aetc.randolph.af.mil/se2/torch/back/2003/0305/runway.htm
GROUND CREW’S INATTENTION LEADS TO UNMANNED AIRCRAFT CRASH
According to an Air Combat Command accident investigation report
released last month, the primary cause of the accident was the
ground crew’s inattention to the aircraft’s altitude.


http://www.globalsecurity.org/intell/systems/predator.htm
As of 31 October 2001 the Air Force had received a total of 68 air
vehicles, and had lost 19 ...
The operator has the camera pointing out the front of the plane,
but he really has lost a lot of situational awareness that a
normal pilot would have of where the ground is and where the
attitude of his aircraft is.


http://www.reviewjournal.com/lvrj_home/2002/Oct-31-Thu-2002/news/19962521.html
May 17 crash of unmanned spy plane blamed on human error


http://www.af.mil/news/Feb2001/n20010202_0157.shtmlOfficials

02/02/01
Officials release Predator RQ-1L Predator accident report
the accident resulted from operator error.

... the pilot -- who flies the aircraft from a ground control
station -- inadvertently cleared the primary control module's
random access memory. As a result, the Predator lost its data link
connection with the ground control station.

Bob Jones

unread,
Apr 27, 2004, 8:02:54 PM4/27/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:dnit805o8fta4g1ep...@4ax.com

>
> FARs prescribe limits on how close to clouds a VFR flight may be, and
> a certificated pilot is trained in recovery techniques.

Where in that report did it mention the flight plan type? You're either
making assumptions again or making another bad comparison.

> According to the report, the pilots' attention became too focused
> on flying the Predator in icing and weather conditions they had
> rarely encountered. The report also cites lack of communication
> between the two pilots during the flight emergency as a cause of
> the accident.

Again, you posted these cites in response to a request for evidence of
"lackadaisical attention to safety". How does this provide such evidence?
For that matter, how is this any different from an IR pilot unaccustomed to
actual conditions suddenly finding himself coated in ice? Human error?
Yes. Additional safety concern? No.

> As you will note, the maintenance personnel did not apply the same
> diligence in maintaining the unmanned aircraft as is required by
> regulation for manned aircraft.

How is this different from certificated A&P's working for, say, Alaska
Airlines? Or any number of similar NTSB reports for GA crashes. You still
haven't made the case that UAV's pose an inherent aviation safety risk.

>> Mechanical failure due to inadequate lubrication.
>
> Yes, and a stripped bolt. That sort of poor maintenance practice is
> not permitted manned aircraft.

"Not permitted," he says. See above. Humans make maintenance mistakes
regardless of aircraft type (manned or not). You still haven't made the
case that UAV's pose an inherent aviation safety risk.

> No. It sounds like you failed to appreciate the operational errors
> committed by the ground based UAV operators and maintenance personnel.

"Operational errors" like the hundreds (thousands?) of similar incidents
committed with manned aircraft. You still haven't made the case that UAV's
pose an inherent aviation safety risk.

> By omitting the citations below, you publicly demonstrate your biased
> thinking:

"Biased", eh? Like you're totally open-minded on this issue. You so full
of BS your eyes are brown.

You keep making the claim that UAV's pose a risk to manned aircraft.
However, even when asked for such evidence, you can't find a single incident
where a manned aircraft was threatened much less harmed by UAV's.

> ...the primary cause of the accident was the


> ground crew's inattention to the aircraft's altitude.

No different from manned aircraft.

> The operator has the camera pointing out the front of the plane,
> but he really has lost a lot of situational awareness that a
> normal pilot would have of where the ground is and where the
> attitude of his aircraft is.

Actually, I did reference this one. See "difficulty landing". You left out
the critical context of your quote. See this (from your link):
"A good number of them were lost due to operator error, **since it is hard
to land the UAV.** The operator has the camera pointing out the front of the


plane, but he really has lost a lot of situational awareness that a normal
pilot would have of where the ground is and where the attitude of his
aircraft is."

[Emphasis added by me.]

Larry, don't bother calling me biased when you're not even *close* to
unbiased.

>
http://www.reviewjournal.com/lvrj_home/2002/Oct-31-Thu-2002/news/19962521.html
> May 17 crash of unmanned spy plane blamed on human error

"Human error" does not equate to "lackadaisical attention to safety." You
still have not proven that UAV's pose an inherent aviation safety risk.

> ... the pilot -- who flies the aircraft from a ground control
> station -- inadvertently cleared the primary control module's
> random access memory. As a result, the Predator lost its data link
> connection with the ground control station.

Human error again, but not "lackadaisical attention to safety." You still
have not proven that UAV's pose an inherent aviation safety risk. (Note:
I'd venture to say this is a bad design feature. The pilot should not be
able to clear the UAV's RAM - especially if doing so can cause the UAV to
lose connectivity with ground control.)


Larry Dighera

unread,
Apr 28, 2004, 1:50:39 PM4/28/04
to
On Wed, 28 Apr 2004 00:02:54 GMT, "Bob Jones" <bjo...@erols.com> wrote
in Message-Id: <0732590b539daf1f...@news.teranews.com>:

>
>> No. It sounds like you failed to appreciate the operational errors
>> committed by the ground based UAV operators and maintenance personnel.
>
>"Operational errors" like the hundreds (thousands?) of similar incidents
>committed with manned aircraft. You still haven't made the case that UAV's
>pose an inherent aviation safety risk.


What if 30% of the manned aviation fleet were to be lost in 10 years
as is the case of UAVs?


Incidently, you haven't contributed any factual information to this
message thread at all.

Larry Dighera

unread,
Apr 28, 2004, 1:51:22 PM4/28/04
to
On Tue, 27 Apr 2004 19:26:56 GMT, "Bob Jones" <bjo...@erols.com> wrote
in Message-Id: <0b55f7a9133746e3...@news.teranews.com>:

>"Larry Dighera" <LDig...@att.net> wrote in message
>news:6bos80pgt5n0m7vb3...@4ax.com
>>
>> I was taught to scan for traffic to the rear of the side to which I
>> intend to turn before making the turn.
>
>That may be, but do you perform this scan in all quadrants in real-time?

If necessary.

>Or only when you need it? My point is you appear to be placing a higher
>standard on UAV's than are placed on existing aircraft/pilots.

That is not my intent.

>>> By the way, where in Part 61 are the vision requirements specified?
>>
>> Here:
>> [really long URL]
>
>Ah, Part *67*, just as I thought. You were either lying or ignorant when
>you made this claim:
>"It's doubtful the...UAV operators' vision systems would function equal to
>or better than that required meet the Part 61 requirements."

Part 61 (§61.23) requires the pilot of an airplane to possess a valid
Medical Certificate, so it indirectly mandates pilots to meet the
vision requirements of Part 67.

Why would you think I was fabricating the pilot vision requirement?
Do you hold an airmans certificate?

>As near as I can tell, the words "vision", "eyesight" and "sight" don't even
>appear in Part 61.

See above.

>Further, you admitted elsewhere in this thread that you
>have no idea what the capabilities of these systems are (yet more
>ignorance).

I believe the UAV capabilities are a bit of a moving target, as its
systems are still under development.

>But why let the facts get in the way of your quixotic quest, eh?

You haven't provided any facts at all, only personal opinion, so you
have no accountability at all to worry over.

Bob Jones

unread,
Apr 28, 2004, 4:41:15 PM4/28/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:317u809600mt575pa...@4ax.com

>
> What if 30% of the manned aviation fleet were to be lost in 10 years
> as is the case of UAVs?

Apples and oranges - again. Military UAV losses have no relation to the
civilian sector. Besides, your original question posed was "who's at fault
in a UAV/Part91 MAC?" Now you're trying to change the argument to "UAV's
losses are too high."

Are you trying to imply something else?

> Incidently, you haven't contributed any factual information to this
> message thread at all.

Neither have you posted any facts to back up your claim that UAV's somehow
pose a risk to Part 91 or any other Part aviation.

You claim that UAV's should not be allowed to fly border patrols due to
their inability to achieve the vision requirements in all quadrants in
real-time as specified in 14CFR Part 61. Only when pressed did you post the
actual requirements - from Part 67 - and then you admitted that you have no
idea what these UAV's are capable of achieving. For all you know, they may
have systems far better than human eyesight.

You are trying to get us all riled up about some perceived safety issue.
However, all you have demonstrated is that there have been significant
losses in UAV's used by the military - many of which have occurred during
landing or in action. Several incidents have been attributed to "human
error" - not necessarily operator error - and none of these incidents causes
me much concern for sharing the skies with them.

You claim I haven't provided any facts and you're right. I don't have
enough to make an educated judgement on this issue. That's why I'm willing
to see the proposals for the operating parameters of the UAV's before I get
concerned, but that doesn't mean I won't call you on your BS.

Face it. You haven't made the case that UAV's pose an inherent aviation
safety risk.

My position remains one of "wait and see". I want to see the operational
plans. Where will they be based? Where will they operate? Will they be
coordinating with ATC? What altitudes will they maintain? What paths will
they take to get to station? Will they have strobes and nav lights to help
VFR visibility? Will they have collision avoidance systems? Other
questions may come to mind after reading their operational plans, but if
these questions are answered to my satisfaction, then I see no problem.

The original quotes you posted mentioned UAV use being considered by
non-military agencies, but you haven't posted any relevant information about
the operational plans or aviation safety record of these agencies.

Nothing you've posted has *any* relation to overall aviation safety. All
you've done is try to stir the pot with implications and innuendo.


Bob Jones

unread,
Apr 28, 2004, 5:10:33 PM4/28/04
to
"Larry Dighera" <LDig...@att.net> wrote in message
news:5mit80lij4v42397e...@4ax.com

>
> Part 61 (§61.23) requires the pilot of an airplane to possess a valid
> Medical Certificate, so it indirectly mandates pilots to meet the
> vision requirements of Part 67.

Do you know the UAV operators won't have valid certificates? If you don't
know the answer to this question, then this point of yours is a non
sequitor, isn't it?

> I believe the UAV capabilities are a bit of a moving target, as its
> systems are still under development.

If you don't know the capabilities (i.e., "you don't have the facts"), then
on what are you basing your opinion?

> You haven't provided any facts at all, only personal opinion, so you
> have no accountability at all to worry over.

You're the one trying to make a case to get the rest of us concerned over
something. Bring your facts and we'll discuss them. All you've done so far
is bring implications and innuendo (i.e., "BS").


William W. Plummer

unread,
Apr 29, 2004, 10:08:14 AM4/29/04
to
ALL planes, manned, remotely piloted or UAVs, are a risk to aviation.
That's why we have the FARs, flight plans, restricted entry to MOAs, etc.
We all operate in the same air. --Bill

"Bob Jones" <bjo...@erols.com> wrote in message
news:f18a23de31e6ba2d...@news.teranews.com...

Tony Cox

unread,
Apr 29, 2004, 11:39:46 AM4/29/04
to
"Bob Jones" <bjo...@erols.com> wrote in message
news:fffd7230c64999a8...@news.teranews.com...

> "Larry Dighera" <LDig...@att.net> wrote in message
> news:5mit80lij4v42397e...@4ax.com
> >
> > Part 61 (§61.23) requires the pilot of an airplane to possess a valid
> > Medical Certificate, so it indirectly mandates pilots to meet the
> > vision requirements of Part 67.
>
> Do you know the UAV operators won't have valid certificates? If you don't
> know the answer to this question, then this point of yours is a non
> sequitor, isn't it?

They may have in the future, but they certainly don't have them now.
Says FAA spokesman William Shumann, "Currently, there are no
FAA regulations dealing with the certification of UAV pilots,
aircraft or (commercial) operators". Nor have I seen any suggestion
that they'll have to be certified in the future.

>
> > You haven't provided any facts at all, only personal opinion, so you
> > have no accountability at all to worry over.
>
> You're the one trying to make a case to get the rest of us concerned over
> something. Bring your facts and we'll discuss them. All you've done so
far
> is bring implications and innuendo (i.e., "BS").

From your point of view, maybe. I've found Larry's contribution
interesting, and certainly something to be concerned about. If you'd
been on this newsgroup longer, you'd realize that it'd be a pretty
empty forum if we just concerned ourselves with "facts".

The most important and worrying aspect of operating UAVs in
the public airspace is that *currently* they will not be operating
under the same constraints as the rest of us. Why? Well for one
thing the entire optical system through camera to operator
isn't certified to private pilot standards.

Of course, you may well be right (as in another post) that "For


all you know, they may have systems far better than human

eyesight". Hell, for all I know, the old duffer at the airport
flying without any sort of medical may well have the best
vision on the planet. But this just doesn't cut it. You can't
'self certify' for vision, and suggesting that this is OK if the "old
duffers" just happen to be the UAV manufacturers is a
significant departure from existing practice. It's not BS, it's
a very valid concern.

Bob Jones

unread,
Apr 29, 2004, 9:43:56 PM4/29/04
to
"Tony Cox" <t...@coxrt.com> wrote in message
news:6l9kc.17327$e4....@newsread2.news.pas.earthlink.net
>
> ...It's not BS, it's a very valid concern.

That's where I differ: I tend not to get very concerned until there are
*facts* to be concerned about.


Dylan Smith

unread,
Apr 30, 2004, 3:37:11 AM4/30/04
to
In article <6l9kc.17327$e4....@newsread2.news.pas.earthlink.net>, Tony Cox i
wrote:

> duffers" just happen to be the UAV manufacturers is a
> significant departure from existing practice. It's not BS, it's
> a very valid concern.

It sort of begs the question - if UAV pilots are going to need to be
essentially PPL standard for medical etc. why bother? Why not just send
the guy up in a Cessna 172 to do his patrols instead? It'd be much
cheaper to stick the man in an existing aircraft that you can buy off
the open market for low (for Government) cost if you're going to need
the man to fly a UAV remotely, anyway.

I see an advantage to using UAVs for reconnaisance over enemy territory.
But over your own country, the only point to UAVs I can see is research
and training the recon operators - which can all be done in a MOA.
--
Dylan Smith, Castletown, Isle of Man
Flying: http://www.dylansmith.net
Frontier Elite Universe: http://www.alioth.net
"Maintain thine airspeed, lest the ground come up and smite thee"

Tony Cox

unread,
Apr 30, 2004, 8:53:50 AM4/30/04
to
"Dylan Smith" <dy...@vexed3.alioth.net> wrote in message
news:slrnc940h...@vexed3.alioth.net...

> In article <6l9kc.17327$e4....@newsread2.news.pas.earthlink.net>, Tony Cox
i
> wrote:
> > duffers" just happen to be the UAV manufacturers is a
> > significant departure from existing practice. It's not BS, it's
> > a very valid concern.
>
> It sort of begs the question - if UAV pilots are going to need to be
> essentially PPL standard for medical etc. why bother? Why not just send
> the guy up in a Cessna 172 to do his patrols instead? It'd be much
> cheaper to stick the man in an existing aircraft that you can buy off
> the open market for low (for Government) cost if you're going to need
> the man to fly a UAV remotely, anyway.

Indeed! But when I suggested that earlier, Bob objected
because it'd send up the price of used 182's! (Owning one
myself, I don't see a problem...)

From the accident reports Larry posted, each of these UAV's cost
us about $3.3 million & need a crew of 7 to keep in the air. That's
10 brand new 182's -- 13 or 14 72's.

>
> I see an advantage to using UAVs for reconnaisance over enemy territory.
> But over your own country, the only point to UAVs I can see is research
> and training the recon operators - which can all be done in a MOA.

There's a good sized MOA over most of Death Valley. They can
pretend the occasional hiker is Bin Laden. I'm with you. I can't see
any reason for operation in the NAS unless it is a "nose under the
tent" issue.


Larry Dighera

unread,
Apr 30, 2004, 10:14:08 AM4/30/04
to
On Fri, 30 Apr 2004 12:53:50 GMT, "Tony Cox" <t...@coxrt.com> wrote in
Message-Id: <y%rkc.84$Hs1...@newsread2.news.pas.earthlink.net>:

>> I see an advantage to using UAVs for reconnaisance over enemy territory.
>> But over your own country, the only point to UAVs I can see is research
>> and training the recon operators - which can all be done in a MOA.
>
>There's a good sized MOA over most of Death Valley. They can
>pretend the occasional hiker is Bin Laden. I'm with you. I can't see
>any reason for operation in the NAS unless it is a "nose under the
>tent" issue.

MOAs are Joint Use airspace where military flights share the airspace
with civil aircraft. Personally, I would prefer that any UAVs
operating there comply with the same federal regulations to which I
must adhere, such as the pilot(s) being certificated to meet
regulatory standards and medical requirements including vision,
see-and-avoid responsibility, _personal_ responsibility under _civil_
and _criminal_ law for the consequences of any damages caused, ...
Holding the ground based UAV pilot(s) personally responsible for any
damage done by their UAV operations might reduce any attitude of
remote anonymity they may feel by not having their bodies subject to
the same catastrophic MAC consequences faced by airborne pilots.
Without personal accountability, UAV operators would have a virtual
license to commit murder/manslaughter with impunity.

(Take for example the irresponsible F-16 military flight leader,
Parker, who led his wingman into a high-speed, low-level, fatal MAC
with a Cessna-172 pilot in 2000. Although he chose to descend into
Class B and C airspace at nearly twice the 300 knot FAR limit for his
aircraft type below 10,000', without establishing communication with
Air Traffic Control as mandated by FAR, rather than face third degree
murder charges as would have been brought against a civilian, the
military found that a verbal reprimand without lost of rank nor pay to
be appropriate punishment for his recklessly irresponsible conduct.
http://www.ntsb.gov/NTSB/brief.asp?ev_id=20001212X22313&key=1)

I would also prefer my government exercise frugality with my tax
dollars, and choose the most effective method of boarder patrol
relative to its cost.

--

Irrational beliefs ultimately lead to irrational acts.
-- Larry Dighera, LDig...@att.net

Bob Noel

unread,
Apr 30, 2004, 6:28:58 PM4/30/04
to
In article <y%rkc.84$Hs1...@newsread2.news.pas.earthlink.net>, "Tony
Cox" <t...@coxrt.com> wrote:

> Indeed! But when I suggested that earlier, Bob objected
> because it'd send up the price of used 182's! (Owning one
> myself, I don't see a problem...)
>
> From the accident reports Larry posted, each of these UAV's cost
> us about $3.3 million & need a crew of 7 to keep in the air. That's
> 10 brand new 182's -- 13 or 14 72's.

well, if they buy new ones, that would eventually increase
the supply of used ones (and they tend to be properly
maintained). This would be a Good Thing.

:-)

--
Bob Noel

0 new messages