Wireless links to be trialled in Gulfstream flight controls Future generations of jet aircraft might use short-range wireless technology to connect cockpit controls to ailerons, elevators or rudders. Flight International reports that the new fly-by-wireless concept will soon be trialled by biz-jet maker Gulfstream, using an unspecified radio technology to control a spoiler on a company …
no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, NO!
Please, stick to having things WIRED UP in planes.
You KNOW it makes sense.
Flight Data Recorder - audio analysis
"Ladies and genlemen, this is your Captain speaking. Is there an IT guy onboard? We appear to be having a little problem with our wi-fi here..."
(recording continues but all further conversation obliterated by continual screaming of passengers)
this makes sense. The problem is going to be figuring out whether the Wireless or Wired/fibered links have priority.
Give the fibre links priority and lots of random flashes as the split end flails in the open air could cause even more problems- or at least tie up the controls while the actuator figures out that its just noise it's reacting to.
Give Wireless the priority and you'd run the risk of laptop-armed Terrorists taking over a plane with Excel and a well-crafted VBA script.
Still, overall a good idea if properly executed. Thumbs up!
So you trust low power, flaky, unencrypted GPS signals to guide the aircraft accross oceans and direct it's landing but not short range encrypted/checksummed/validated radio signals inside the plane?
There are lots of high reliability protocols used in radio controls for cranes, mining machine tools etc.
Correct, as a backup it's a great idea, cut fibre isn't going to be an issue, because each command sent should really be signed/checksummed anyway, so any noise won't checksum properly. Backups should be as diverse/different from primary links as possible - No good having a diverse system that is susceptible in the same way as the primary, especially true for flight computers, no good having two/three identical systems, as a bug in one, will be in the others as well. Same thing on a physical level, no good having a 2nd set of cables, because they would be susceptible to the same problems as the primary i.e. being cut.
The issue is using something to attempt to gain control after the power linkage has been cut.
Back in the day, this would mean you lost pressure in your hydraulics.
Today that means loss in your "wire". Tomorrow, optical fiber.
The interesting thing is that you can build redundancy in your optical links, but even in the event of a catastrophic failure, I doubt that wi-fi or its equivalent will help much.
As to the GPS signal, encrypted or unencrypted, you have the same issues regarding accuracy. GPS (Civilian) is accurate enough. Not flaky at all. But no, its not used to direct "the landing". (BTW, GPS is very accurate when used in farming. But there you have a base station.)
I think that its interesting that they would try a short haul wireless concept. Definitely not my first choice of a back up.
Tux because you know the network routers *will* use a flavor of Linux. ;-)
If done right this makes a lot of sense in the same way that remote office sites sometimes have wired network connections and a wireless solution in place as a backup in case a cable gets cut.
As planes now use fly by wire technology the opportunity exists to migrate the analogue controls to digital data packets which can be correctly checksummed and routed around redundant paths in the same way that routed networks can. In this scenario you can have a correctly authenticated encrypted channel from the cockpit to the control surfaces which can travel over the wireless links or through several redundant fiber runs in the fuselage. In this case a terrorist with a laptop and the correct hardware for the airplane's wireless communications (let's face it, nobody would be stupid enough to use 802.11n here) would be able to do little more than jam (or DoS) the signal which would have to be combined with a physical cut of the redundant cable paths to do any real damage.
Cranes and mining tools generally speaking don't carry passengers. Plus the navigation can be backed up by radar and radio triangulation, not to mention the pilot and flight crew. Just as they used to do prior to GPS.
High reliability is one thing, I wonder just what level of redundancy they intend building into these systems. Would they be more expensive overall than wires? Would they be more reliable than wires?
If used as Gulfstream propose, I'd imagine that the'd have way more than 'loss of control' to contend with if they blew a rotor on one of the turbines. More like a loss of the majoriy of the tail section. Frankly, if they're worried about that, I'd say they had an engine reliability issue foremost, and I'd be concentrating on that instead!
This is what ground based engineering has been doing for the last 15 years.
SCADA and PLC automated system controls use a system-wide network for sensors and control signals. Some are even Ethernet compliant.
This offers many benefits over hub-spoke wired systems. Using a wireless bridge across possible failure points is good engineering.
This is quite far-sighted for the usually conservative aviation sector.
No, I don't trust them to fly the plane. I trust them to tell a human pilot where to fly the plane. I also trust all the land-based radar tracking stations to report on the plane's position, thereby ensuring that the low-power signals are indeed leading the pilot on the right course.
As for fiber security, obviously if you run the wires and fiber through the same path, cutting one will cut the other, duh. In a plane there are 4 cardinal points. I expect one technology to be routed through one point, and the other tech to be routed on the opposite side - but I'm no expert in avionics.
I guess WiFi might just be acceptable as a backup, but only if there is no electrical interference in the surrounding atmosphere. If your engine has blown up violently enough to shred the fuselage, you're a goner anyway, so I don't believe that WiFi would help in that situation. Again, I'm no expert in avionics, so maybe it is a good idea.
If you can't hack it with Excel and a VBA script, that is.
A lot of you seem to miss the point that this is intended as a backup system, which by all means is a bloody brilliant idea!
It could be the difference between a damaged plane going down taking everyone with it, or a damaged plane still able to maintain some landing control thanks to the wireless backup.
1. People, on average, aren't as stupid as is frequently imagined by some Reg readers.
2. This will not reach production until it is thoroughly tested.
3. For those who can't accept any argument unless it is accompanied by a negative 'people and corporations are governed purely by self-interest' there is the 'truth' that they will not want to jeopardise their profits by risking having a large number of aircraft falling out of the sky.
Smarten up.
"... at least in the near term" - that's the scary part. As a backup - no problem because it is better than fly-by-cut-wire.
But as a primary control - no way. It is too easy to ruin the signal. Solar flares, really dim hackers, politicians with wireless network blockers, electronics-frying bombs, directional wifi disruptors, and lots of other potential problems.
Actually, thinking about it, maybe it would be a good idea to test the prototype on Air Force One, depending on who wins the upcoming erection...
Paris because of the (probably intentional) spelling mistake...
Because of solar scintillation, I doubt wireless will ever be used as a primary means of flight control. However as a backup means, it does make sense for controls which are already electronically controlled using servos etc.
Wouldn't do much for systems using pneumatic or hydraulic means. Since a cut in those systems lines would mean a loss of the medium itself.
Just how many aircraft are built with electronic servos actually moving the flight surfaces? I believe right now, most are still assisted via hydraulics.
I assume you mean with duty-free souvenir nail-clippers that you can still get airside...
Really bad idea. Depending on your intentions, you don't even need to control the plane once the primary (for the moment) control-method is disconnected if you're quite happy to simply disrupt the relevant bandwidth resulting effectively in a pilotless plane...
If only these flights could have had a "backup" system like the one proposed:
http://en.wikipedia.org/wiki/United_Airlines_Flight_232
http://en.wikipedia.org/wiki/Japan_Airlines_flight_123
http://en.wikipedia.org/wiki/Turkish_Airlines_Flight_981
That's nearly 1000 deaths that would probably have been prevented if there had been an alternative means to control the control surfaces and engines than a "hard link".
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
