if you noticed where the drone footage was from, it was above Ft Dennison. right in the middle of fireworks restricted zone, i am pretty certain, that CASA would have only allowed it to fly there, after filing a flight plan, and risk assessment forms, probably done after the last NYE show.. so, even if it was hit by a firework, it would have crashed into the harbour a safe distance from everyone.. even those flying over the fireworks on a harbour scenic 2 would be in no danger, min 2100ft and all.

There didn't seem to be all that many scenics on the night, the odd chopper and the ApplicancesOnline blimp had been flying for much of the day and into the night.

Unfortunately I don't know him that well but the UAV operator was Rotorworks. I would be certain they would have all required permits and licences. If you want to see what I could see look up "Surf 2 Summit Media" on Facebook.

Ultralights, from the photos there, you're spot on. They were right at Fort Denison. Looks like they were pretty well set up for the night!

Yes Lookleft, that's my impression of the risk assessment too.

So the risk assessment included historical data or research of the performance of a UAS after struck by fireworks that indicated it would crash straight down?

Maybe it did. Like I said, I'd love to read the risk assessment on this one.

Drones could do anything when hit by fireworks... It's electronics after all and a knock could upset the gyros and it could spear off in any direction and into anyone. All the failsafes in the world may not stop it... Refer to ATSB report for loss of control of a UAV at the Melbourne MCG.

But, are they safer than a single engined Squirrel overhead at night? Probably.

On that basis, no "first of type" or "first time this has been tried" would ever get off the ground, correct?

Or would you set up a firework display or a series of displays in the middle of nowhere and fly a UAV through the display hoping that it would get hit to determine how it would react?

How many times would you need to do this to get sufficient "historical data" and how many hits would you need to determine that the struck UAV would not do other than crash straight down?

What a stupid question! First of type go through rigorous design and testing before they even take-off.

The statement suggests that this experiment was done with a lot of fingers crossed.

G'day 601. I'd want to know how well the UAV flies after having one of its rotors smashed and the thing turned on its side. Maybe it'd plummet straight down. Maybe it wouldn't. I'd want to know how well it recovers from a hit to the electronic control box. IE when the 'link' is cut. Then do that with the rotor out. Then a couple of rotors out.

We aren't talking about a first flight of a type. We're talking about flying a known type into a completely hostile environment. It's basically a combat zone but one with hundreds of thousands of people hanging around as close as possible to an exclusion zone that was designed for fireworks, not errant UAVs.

Fly just above the fireworks, beside them, heck, even from below and to the side. Through them though?

By type I was referring to the operation NOT the aerial vehicle.

These multirotor autopilots sacrifice altitude control for the inner control loops which provide aircraft attitude stability through differential thrust. A loss or partial loss of power to one or more motors in an autopilot mode will cause the remaining motors to reduce power and the aircraft will descend. Of course the most likely scenario is the machine immediately loses stability and crashes.

The wreckage will be found very close to where the original issue occurred in either case.

The likelihood of a "fly away" event due to getting hit by fireworks is so remote it does not need additional control measures to reduce that risk further.

Have a read of the ATSB report on the Melbourne MCG drone as it lost control (and had a fly-away occurrence) possibly due to saturated frequency airwaves due to the high number of mobile phones due to the footy match. How many phones etc were there for NYE???

i thought it was television transmissions from vans parked below it that was teh cause of that one? as mobile phones are ona differeing frequency, but television is pretty close.

Right you are UL, hence my use of the term "etc" to cover the full spectrum of frequencies :-)

From ATSB report summary, my bolding.

The operator’s investigation into the accident concluded that radio frequency interference was the most likely cause of the accident.
...
The operator acknowledged that further testing and analysis was required before the primary cause of the accident could be confirmed beyond doubt.

Given the profile of this incident, it would have been very beneficial to the sector for the ATSB to have conducted or arranged an independent investigation into the cause of this accident and based the report on actual evidence.

The availability of autopilot telemetry would show very specifically what went wrong.

Investigation: AO-2015-112 - In-flight break-up involving a DJI S900 remotely piloted aircraft, at Toowoomba, Qld on 19 September 2015

Although short, this one makes for some interesting reading. This one could have ended very differently!

Probably NOT.

In the period 2000 to 2013 there were over 150 crashes of helicopters and a few fixed wing, whilst engaged in aerial filming or photography.
Whilst there were scores of crew injured or killed in these accidents, no member of the public was reported injured or hospitalised.

Yet filming drones have already injured bystanders and members of the public.

So whilst drones are far safer than helicopters for their crew they are not yet safer for the public.

I'd agree that unless a independently powered kill switch with a dedicated receiver was used, then there was a risk of partial damage by fireworks and loss of control that could send the craft off into a crowd, rather than straight down.

Note: The incredible record of filming pilots keeping the public safe was broken last year when a news chopper fell from a rooftop helipad and crashed into a car, killing its driver.


Mickjoebill

Does this not prove that there are huge areas available in which to crash even sizable aircraft with little risk of hitting a person?

Drone operators both recreational and professional must continue to improve performance in managing the risks, but let's not overstate the real risk to the public.