View Single Post
Old May 28, 2011, 11:55 PM
Daemon is offline
Find More Posts by Daemon
Registered User
Daemon's Avatar
Lakewood, Colorado
Joined Aug 2002
30,717 Posts
Originally Posted by Mel Duval View Post
OK, so let's talk numbers. The statement was made that " If the true odds of collision in class G airspace are say 1 in 100 billion", then a number of extrapolations and relations were made. Some context may be instructive on this number. The FAA (and other civil aviation agencies) have stated on a number of occasions that the analytical incident probability of occurrence they are looking for sUAS is between 10E-6 and 10E-9 (one in one million to one in one billion operating sorties/hours) (Patrick Egan, catch me here if I am going off course). Turns out the general aviation accident rate has been hovering around 6 in one hundred thousand hours (6 x 10E-5) so they are not meeting their desired numbers there for GA.
True, but their accident rate is higher for all the reason I mentioned.
They are forced to fly to and from the same fixed locations and naturally
concentrate in specific corridors.

A group wanting to launch a weather balloon carrying a glider did some pretty fair calculations along these lines. See this site: Their numbers for what appears to be a relatively busy General Aviation area turned out to be about one in 600,000 sorties. The calculations seem fairly reasonable and are WAY less than the number suggested up top.
Yes they specifically chose very busy airspace to give themselves the worst odds possible, but followed up with.
"This leaves a chance per daylight hour of intersecting a small randomly placed aerial object at 1 in 300,000 flight hours, assuming no effort at all is made to avoid airports, practice areas, or traditional waypoints.

But GA mid-airs overwhelmingly occur in clear daylight, in the vicinity of airports, and either in the approach path to landing, or at the discrete altitudes light aircraft often fly at (1000', 1500', 2000', etc). In addition, flight tracks generally begin and end at an airport or common waypoints. These facts tell us that by actively avoiding GA airports and practice areas, the risk can be brought substantially below the above value."
(emphasis mine).

They ultimately come to the same conclusion that I said above (and I've never seen this
specific report before). That to lower the risks significantly you simply move away from the
high traffic areas, which we generally already do.
BTW, they reference the MIT study, but their link is broken. Here's a working link.
Notice that in their mid-air collision risk analysis, once you move away
from highly traffic'ed areas the odds do drop into the 1x10-9 to 1x10-10 range and even less.
Look at all that blue and white area of the first map on page 15.
It really is not that hard to know where full scale aircraft operate
the most, and avoid those areas. It would not be impossible to create regulations
which say to do that, or present different rules for flying in lightly trafficked areas
as currently exist for various types of ultralight/HG/PG aircraft.

This is how the government (and everybody else, really) evaluates risk. So if you want to minimize risk, one way would be to mandate that all FPV flying be done with < 4.0 lb. foam pusher aircraft that fly less than 35 MPH. Then control where they can fly and how high. Now THAT would reduce risk by controlling the probability and the consequence.
That's at least a start.

And for those who say that the FAA cannot possibly enforce the upcoming regulations, they won't have to. When someone flying outside of the approved parameters crashes into something on the ground and damages something/injures someone or has a midair out side the field limits doing the same, the LAWYERS will enforce the regulations by taking every thing of value from the offender as soon as they take one look at the regs. And they will......
Which is no different than today, regulations or not, and the reality is that
light model aircraft crash on top of or into things all the time, and little or no harm is
done and operators are rarely sued to their ruin.

Daemon is offline Find More Posts by Daemon
Last edited by Daemon; May 29, 2011 at 12:22 AM.