On Mon, Jun 25, 2012 at 6:43 PM, lists@lazygranch.com wrote:

I think it is.  Currently the military uses GPS guided drones put the
article says they will see more and more used, even by companies like Fed
Ex.  I don't think I deliver this will happen soon but it might.  The
article says that these new drones will be susceptible to GPS spoofing.

But there is a simple "fix" that to me seem obvious.  When you design a
nab system for a drone they should use inertial nav.  You can't spoof an
IMU.  But the cheap IMUs drift and need GPS updates.  So each time you
update the INU to do a sanity check on the GPS and see if it is within the
drift range of the IMU.  If not you assume the GPS is being spoofed and
continue using the INU data.

There are other backups too.  One is commercial radio broadcast stations.
You can measure doppler to several of them.  And let's not forget about
video cameras that can see the ground or up to stars.  lastly there is
common sense that could be programmed into a navigation system, for example
if flying a straight course you'd expect the sun angle to remain constant
and the horizon to remain level

Same with any human pilot, truck driver or hiker, all of these people know
it is dumb to rely only on GPS.  Everyone cross checks it with other nab
sources.

--

Chris Albertson
Redondo Beach, California

On 6/25/12 7:11 PM, Chris Albertson wrote:

And of course, this is the way almost every autopilot/nav system out
there works..

You use IMU+GPS... GPS is long term, but crummy in the short term; IMU
is good short term (e.g. to stabilize flight path), but crummy long term.

Not to mention that spoofing GPS is actually fairly hard to do,
reliably.. you have to have an internally consistent set of signals and
observables that seamlessly connects to the original natural set and
then walks off.

jamming is easy, spoofing is hard.

I would also expect that these things will very quickly go to L1/L5 for
"safety of life" applications, and spoofing 2 frequencies is just that
much harder.

Whether it's spoofing or jamming, domestic drones are becoming ubiquitous,
because they are just so tempting, and sooner or later one is gonna crash
onto a populated area, either by accident or deliberate mischief.

A piloted aircraft may be able to avoid hitting a school; a drone may not.

-John

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

On 6/26/12 3:38 PM, J. Forster wrote:

That is the significant problem with non-government UAVs.  All fine to
run them over the desert on the southern border or out over the Mojave.
By and large, UAV failures, as you note, don't have the option of
doing a Great Santini.

The  (catastrophic) failure rate of UAVs is something like 100 or 1000
times higher than for military piloted craft, which in turn is something
like 100 or 1000 times that for civilian craft.

I did some calculations last year, and if Los Angeles decided to put up
a UAV 24/7 to replace things like helicopters, we could expect a crash
into the city about once a week.

The MQ-9 Reaper and RQ-1 Predator have a reported Class A mishap rate of
about 10 per 1000 flight hours...  Class A = >$1M in damage or death..
bear in mind that if a $500k drone augers in out in the desert, that's
not a Class A mishap.

So, 1 year is about 8760 hours, so we could expect 87.6 Class A
mishaps/year if the LAPD decided to fly the current flavor of UAV.  Yes,
that would create some interesting news stories.  How long til we see a
tailfin with LAPD sticking out of an elementary school a'la Cerritos.

For comparison, in around 2000-2005, the commercial accident rate was
about 0.01 per 100k hours.  The Air Force reported about 1 per 100k
hours.  General aviation is 10/100k hours.  (these are non-specific
"accidents", so they aren't directly comparable to Class A mishaps)

There's a great report from MIT on this.. google for Weibel ICAT report
UAV safety

IMO, your failure rate estimate does not include the probability that some
people might not like being spied on by UAVs.

I can easily see a market for ground based GPS jammers, especially, in the
more rugged, fertile, and inaccessible areas of California.

YMMV,

-John

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

If the GPS is jammed, the UAV goes into a failsafe mode. That can mean a lot of things, but often just orbit. I FOIAd a crash near Creech that sounded according to newspaper account as to crashing in "free territory". It failsafed into the side of a mountain located on restricted territory.

The deal with UAVs used in cities is they will probably be controlled via  terrestrial link rather than satellite. Satellite control can be troublesome.  Weak signals and significant latency.

-----Original Message-----
From: "J. Forster" jfor@quikus.com
Sender: time-nuts-bounces@febo.com
Date: Tue, 26 Jun 2012 16:42:16
To: Discussion of precise time and frequency measurementtime-nuts@febo.com
Reply-To: jfor@quikus.com, Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: Re: [time-nuts] Spoofing GPS

IMO, your failure rate estimate does not include the probability that some
people might not like being spied on by UAVs.

I can easily see a market for ground based GPS jammers, especially, in the
more rugged, fertile, and inaccessible areas of California.

YMMV,

-John

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

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
.

On 6/26/2012 7:57 PM, lists@lazygranch.com wrote:

If the GPS knows it has been jammed, the UAV goes into a failsafe mode.

There, fixed that for you.

On Tue, Jun 26, 2012 at 3:59 PM, Jim Lux jimlux@earthlink.net wrote:

But they could be made very safe for only a little bit of money.  Say you
add a rocket deployed parachute triggered by ground proximity.  These kind
of chutes are made large enough for light aircraft.

Lots of other things to do for safety like a video camera that is monitored
and then you'd know in a minute if the aircraft was going the right way.

Chris Albertson
Redondo Beach, California

I have been around military jamming. The GPS goes to zilch. It isn't a soft degradation.

-----Original Message-----
From: Mike S mikes@flatsurface.com
Sender: time-nuts-bounces@febo.com
Date: Tue, 26 Jun 2012 20:15:03
To: time-nuts@febo.com
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: Re: [time-nuts] Spoofing GPS

On 6/26/2012 7:57 PM, lists@lazygranch.com wrote:

If the GPS knows it has been jammed, the UAV goes into a failsafe mode.

There, fixed that for you.

time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Report title:

Safety Considerations for Operation of Unmanned Aerial Vehicles in the
National Airspace System
Weibel, Roland E; Hansman, R. John

Link is here:

http://dspace.mit.edu/handle/1721.1/34912

2005 so a bit behind current state of art but what municipalities are going
to pony up for the latest technology...

Dave

On 6/26/2012 9:53 PM, lists@lazygranch.com wrote:

Whoosh. The (off-topic) discussion is about civilian GPS, as used by
civilian drones. I take it you didn't read the linked article from the
OP, which described the successful operation of a $1000 spoofer.

On 6/26/12 4:42 PM, J. Forster wrote:

The GPS antennas on the UAVs face up, and have a (not very deep) null
facing down, so those rustic farmers are going to need a fair amount of
power facing up...

(and besides, why do it outside, when you can inexpensively rent large
industrial concrete tilt-up buildings with water, heavy power, and a
loading dock..)

And, as far as remote sensing goes.. commercial overhead imagery from
aerial photo and satellites (SPOTimage, etc) is sufficiently good to
detect this kind of thing (as well as things like unpermitted swimming
pools on which property tax is not being paid).

On 6/26/12 5:51 PM, Chris Albertson wrote:

No, that doesn't make it safe.. it still fails (engine failure is most
common) it just potentially allows you to crash somewhere less
obnoxious.  I suspect that the overall system reliability of a UAV is
substantially lower than a military jet.  They're not doing things
like multiple redundant communication links on different bands, or
redundant control systems or redundant anything (all of which commercial
aircraft have)..

Dropping a UAV by parachute onto a school is not quite as bad as
augering into a pre-school, but not by much.  Or in the middle of the
freeway during non-rush hour.

When I was getting my pilots license, I used to have bad dreams about
having an engine failure flying along PCH in Malibu at 1500-2000 ft, and
trying to decide whether to land on PCH (tons of wires) or the beach
(tons of people). We used to have discussions about whether it's better
to land on the freeway going with traffic (lower closing velocity but
you're coming up on people from behind) or going against traffic (people
will see you coming and hopefully dive for the shoulder).

The smaller path loss from the ground to a UAV v. UAV to satellite easily
trumps the front/back ratio of most all antennas.

-John

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

Mount a GPS antenna on the bottom of the UAV. If you get strong signals
from that antenna, assume information assurance has failed.

There are countermeasures, and of course counter-counter-measures.

Here is a photo of a predator spying on me:

On 6/26/2012 8:16 PM, J. Forster wrote:

Interesting this idea: mounting a bottom GPS antenna to check for
unexpected signals... and you don't need a complete GPS receiver maybe only
a sort of selective field strength meter or something like the codeless GPS
receiver used by Vaisala on their radiosondes.

On Wed, Jun 27, 2012 at 6:45 AM, gary lists@lazygranch.com wrote: