provided by: 
In the wake of the small-plane crash into
a Manhattan apartment building last week, politicians and the public
have focused on the potential dangers of unmonitored small-plane
traffic around New York City.
But
a technology question arises: Why don't these planes--which aren't
required to carry "black box" voice and data recorders--at least
required to save their GPS position information, for more accurate
accident reconstruction?
The
plane that crashed, killing N.Y. Yankees pitcher Corey Lidle and flight
instructor Tyler Stanger, was last seen by conventional radar units
about a quarter-mile away from the accident site, in the middle of a
U-turn, at an altitude of about 500 feet, according to the National
Transportation Safety Board. That leaves a lot of data missing about
what happened in the final plunge before it exploded against the
building.
Inexpensive equipment could fill that void.
John Hansman,
professor of aeronautics at MIT and director of MIT's International
Center for Air Transportation, discussed this and other crash-related
issues with Technology Review.
Technology Review: What is the difference between radar and GPS in terms of the ability to record flight data?
John Hansman:
Radar registers a plane's position every time the radar sweeps by,
which is typically [every] four seconds. GPS typically calculates
coordinates once per second. And at low altitudes, radar can be
shielded by tall buildings. Radar doesn't actually measure altitude--it
measures position--and the plane responds by reporting its altitude to
the nearest 100 feet. GPS data on the plane can be accurate to a few
feet. And it can be stored in the GPS unit on the plane. There is also
an emerging system called Automatic Dependant Surveillance-Broadcast
(ADS-B) that would allow the GPS data to be broadcast and received by
air-traffic controllers.
TR: What's the significance of that radar/GPS difference in terms of this crash?
JH:
With GPS, you'd be able to better reconstruct what actually happened.
Was he descending? Was he descending at a rate that was equivalent to
having lost an engine? Was he turning? There's a lot of stuff they will
get out of the forensics--how the propeller bends on impact depends on
whether the plane's engine was working. But from GPS you could get more
information on how the airplane was maneuvering prior to impact.
TR: Did the plane that crashed, a Cirrus SR20, carry any GPS equipment?
JH:
Yes, it had a GPS that fed data into a multifunction moving map display
(MFD). GPS trajectory storage is a function included in some GPS units.
I am not sure what unit was on this aircraft, so it is hard to say
exactly whether it had a GPS record. Apparently, in the NTSB briefing
they did indicate that they had recovered a memory chip off the MFD.
But I don't know what data was recorded on the chip and whether the
data would have been wiped out during the crash. There is no
requirement for flight-data record for general aviation, but
investigators will take advantage of any evidence they can get.
TR: Do you think the FAA should require small planes to record their GPS data? It wouldn't be so expensive, would it?
JH:
Take your garden-variety, hand-held GPS unit, and it can show you the
track of where you have been. It doesn't take a lot of memory to record
the coordinates of recent flights of an airplane. It has become very
common to archive the trajectory data in memory. We do this sort of
thing with cars now. We are naturally moving into an era where most
accidents will be more accurately diagnosed with data residing in
electronic systems.
What
is happening now is there is a de facto move to have that capability
onboard, because it is so cheap. However, if you make it a requirement,
it would get complicated. You would require not only the memory, but
some level of protection so the memory doesn't get burned, so it's
survivable.
By David Talbot
Read article at techreview.com