PDA

View Full Version : Damage-Tolerant Flight Controls


ORAC
16th Nov 2010, 08:02
Looks like the sort of software that could usefully migrate across to manned aircraft.

AW&ST: Damage-tolerant Flight Controls Demonstrated (http://www.aviationweek.com/aw/generic/story_generic.jsp?channel=defense&id=news/asd/2010/11/15/12.xml&headline=Damage-tolerant%20Flight%20Controls%20Demonstrated)

Rockwell Collins has demonstrated its damage-tolerant flight controls can increase the safety of an in-production unmanned aircraft, with a series of flights on the U.S. Army’s AAI RQ-7 Shadow tactical UAV. The damage-tolerance software was loaded onto the Shadow’s Rockwell Collins-supplied Athena flight-control computer and several flights were conducted to demonstrate safe recovery and landing after various flight control and engine failures, including losing part of the wing.

The flights completed Phase 3 of the U.S. Defense Advanced Research Projects Agency’s Damage Tolerance program. Under previous phases, Rockwell Collins demonstrated safe recovery and landing of a subscale F-18 model after damage including the inflight ejection of 60% of the right wing and 30% of the right horizontal and vertical tails.

In the Shadow trials, the UAV automatically recovered controlled flight and landed itself after both neutral and hard-over failures of ailerons, elevators and flaps, says David Vos, Rockwell’s director of control technologies and unmanned aircraft systems. The outboard 20 in. of one wing—about 10% of its span —was blown off in flight and the UAV automatically recovered and landed, he says. Simulated engine failures and a complete shutdown were also conducted, and the aircraft recovered, selected a landing site and glided to a safe touchdown, Vos says.

“We took a production Shadow off the line, went out with new damage-tolerant flight-control laws and flew five or six flights in a few days, with no multi-month qualification program,” he says. “This was an important milestone. The next step is how we make this a standard solution.”

A 2004 study of Shadow accidents showed 50% were caused by failures on the aircraft, and in 2005 the Army launched a program to improve the reliability of the UAV’s engine. With damage-tolerant controls, Vos says, “When you do have an engine problem, the aircraft lands itself safely.”

The locations of suitable landing sites are loaded into the flight-control system before takeoff. “It always knows where the nearest safe landing site is, autonomously, even if you lose link with the aircraft,” he says. If the engine fails, the system automatically selects the best landing site and plans a trajectory to the runway, Vos says, continuously calculating wind and maneuvers such as a spiral descent and crosswind de-crab before touchdown.

Although Darpa’s Damage Tolerance program is now complete, Vos says Rockwell Collins is continuing research into adding capabilities and functions to its flight-control software.

Mechta
16th Nov 2010, 16:49
I'm rather underawed by this if the UAV only lost 10% of the wing. Especially when one considers that the A-10 Thunderbolt is designed to fly without an outer wing panel, half tailplane and one fin/rudder assembly. A bomb damaged Douglas DC-3 was fitted with a DC-2 wing panel which made the wing 5 feet shorter on one side, and flew 'ok', and there is of course the Israeli F-15 which made a successful landing with almost a whole wing missing.

http://i817.photobucket.com/albums/zz96/s4racergp/DC25.jpg

The F18 performance quoted sounds more of a challenge.

What is really needed now, is a way of controlling engines and control surfaces after the wiring harness has been damaged. Both the Etihad Airbus A340-600 at Toulouse a couple of years ago, and the recent Airbus A380 incident in Singapore demonstrated that an engine will keep going after the control has been lost, and there's currently not a lot that can be done about it. The A340 was more or less left to run out of fuel!

It shouldn't be too difficult to have a back-up system of a microwave or laser link between the fuselage & engines, so that a crew could at least shut down, or possibly even control an engine without the interconnecting signal wiring. This would obviously need a source of power at the receiver and to power the control on the engine. An RF system would be possible, but that could then leave the system vulnerable to interference.

just-nick
16th Nov 2010, 16:57
Keeping/getting engines shut down isn't just a problem for throttle-by-wire/FADEC systems. I'm sure everyone here is familiar with the runaway Cessna Citation Boat (http://www.youtube.com/watch?v=N1Yf6_MVTck)...

Cheers,

Nick

Mad (Flt) Scientist
16th Nov 2010, 21:21
I suspect the challenge in applying any king of damage tolerant (or self-reconfiguring, because that's what it amounts to) flight control system to a passenger carrying aircraft will be addressing the failure cases.

And, psychologically, there are enough people who don't trust the FBW programming we have today, and that's a known entity which will be identical every time its produced. What these people would make of truly "intelligent" flight control computers ...?

ChristiaanJ
16th Nov 2010, 22:52
"What these people would make of truly "intelligent" flight control computers ...?"

I can imagine some of them pulling them out of the rack and throwing them out of the DV window.

"HAL? Why did you do that?"

CJ

Denti
17th Nov 2010, 03:05
I wonder just how many of those UAVs were lost that the army launched a program to increase reliability in the first place? Not really something that inspires confidence in pilotless aircraft.

Escape Path
18th Nov 2010, 00:51
Not really something that inspires confidence in pilotless aircraft

And yet, there are folks out there in the flying public that believe that we are useless, somehow :rolleyes:

VnV2178B
18th Nov 2010, 07:01
The EECs controlling the Airbus engines, and Boeing's too, have dedicated generators which will provide all the power you need to receive control data as long as the engine is running. This creates the shut-down problem in the first place as if no cockpit signal is received the engine will carry on at last commanded value until the fuel runs out or it gets flamed-out some other way.

Thanks for the picture, Mechta.

I remember this being immortalised in the Gooney Bird song "One wing askew, and yet she flew, the DC two and a half. ..." but never had any real evidence!

VnV

Mechta
18th Nov 2010, 22:54
VnV2178B, you're welcome. Here's the original DC 2 1/2 article: http://www.cnac.org/aircraft02.htm (http://www.cnac.org/aircraft02.htm)