I enjoy watching air crash documentaries. I’m fascinated by the technology available in these airplanes as well as the science involved in discovering why they crash. I’ve noticed that quite often investigators are puzzled by the actions of the pilots on board, and they’re forced to rely on limited data to make their conclusions. They also lack the ability to see exactly how the plane was behaving before it crashed. Some mysteries are solved with the help of a witnesses recording, photographs, or security camera footage. These are answers a cockpit video recorder would easily provide. Why not record what’s going on in the cockpit and maybe mount the equivalent of a police car dash cam?
We record the flight data. We record the cockpit audio. But we don’t record the video? Surely, you can’t be serious.
It begins to make more sense when you realize that flight data recorders record all the flight data and audio on a single loop of heavy gauge wire. The emphasis is on durability and ability to survive a catastrophic event.
There is nowhere near the capacity you’d need to record analog video. (Well, digital, either, but even if there were, you want analog because it’s easier to recover from if the signal is degraded.)
I don’t think so, not anymore anyway. First the audio and flight data are recorded on separate devices (the cockpit voice recorder and flight data recorder respectively.) Second, modern CVR and FDR devices are solid state. This particular model uses flash memory (PDF).
There is a move toward using cockpit video recorders and I think it will eventually come. There is a bit of resistance to it from the piloting community though (much as there was towards CVRs).
[QUOTE=AtomicDog]
A terabyte drive can hold well over 1000 hours of video. I don’t see storage and processing as being an issue.
[/QUOTE]
Find a way to make a spinning drive survive a 3,400g impact and being baked at 2,000 degrees Fahrenheit for half an hour, and then you might have a viable proposition. (Specs are from the data sheet Richard Pearse linked to.)
Before you jump up and say “What about flash memory? 64 gig SD cards are pretty cheap now.”, those data recorders dedicate a whole lot of mechanical and electrical armor to protect 37 meg of data, (yeah, that’s a weird number, but that’s the spec) and the support infrastructure runs on 486DX-33 MHz class computers. Almost Stone Age simple equipment, but it is as rugged as a rock.
Now the question is, what do you aim a camera at? A control panel? Which one? You might need several cameras. Captain or co-pilot? Can you even find a place to put a camera that it will have an unblocked view, regardless of how someone is sitting/leaning/stretching/reaching? And how do you get people accustomed to being videotaped at work? How do you safeguard against footage of someone picking their nose from getting posted on Facebook?
I don’t think the problem with video records is a technical one, but rather a labor/privacy issue. I imagine the next step in this technology is to just have a constant satellite upload of this data in real time to storage on the ground. Technology won’t be the limiting factor in what is recorded.
For commercial aircraft manufactured after Dec 31 2002, the CVR must be able to record the entire flight, or two hours of data, whichever is less (CAR PART VI, 625.33 section II.(3)).
I don’t know much about computers, so perhaps this is irrelevant, but FDRs and CVRs are increasingly solidstate drives. I don’t think the storage media is really an issue for video recording, though perhaps the necessary size, weight and electrical harness might be.
This is a much more relevant set of concerns in the industry today: pilot associations do not want the continuous videotaping of their work for fear of the data being used against them by their employers (because let’s face it, no employee ever follows every rule 100% of the time).
There is also the lack of angles/things to point a camera at in many cockpits; smaller aircraft (CRJ, E-jets) have itty bitty cockpits and there is no single place you could put a camera that would capture everything anyways. I would think you’d also need some high-definition recording, which adds complexity, weight and cost to the system. The combo of recording what buttons are pushed/what settings selected (FDR) and being able to hear what the pilots are saying (CVR) is actually quite good at getting the whole picture across to investigators.
I think we will see mandatory FDR/CVR with more channels being recorded on the FDR before we get video. Older planes and some classes of aircraft don’t have to have both and currently use older, grandfathered systems that don’t record enough…I think the NTSB/TSBC/etc would prefer to retrofit older planes rather than introduce new technology.
I don’t think anyone would be reviewing the footage unless the plane crashed. But just the same, I don’t think anyone wants to see footage of the pilots dying either.
I was thinking more of a camera that would be pointed at the instrument panels, just to see what adjustments the pilots made and to see what instruments may have been malfunctioning. It’s often hard to tell if something was previously broken when you inspect it after impact.
I don’t think a camera pointed directly at the pilots would be that useful.
Oog… I hadn’t even thought of having someone’s last moments on video. As for not reviewing footage, it will be reviewed at least once a year to ensure the device is working properly.
Sorry no cite, but my friends in the defense biz say that military planes record everything - out the window, the instrument readings, and I assume the pilot. They play this back when some component fails which is pretty frequent. Some one said that wheeled vehicles do this also nowadays.
There really isn’t much space in many cockpits that aren’t already being used, that could see the entire instrument panel or that isn’t already provisioned for future upgrades/technology (sometimes airframers plan things weird!)
For the bigger planes that aren’t grandfathered, modern FDRs record upwards of 100 or more data parameters; there aren’t many buttons and settings on glass cockpits that aren’t also being recorded by FDR (anything flight critical is recorded nowadays).
For older aircraft, it would be easier to just upgrade to new FDRs than introduce new technology because you can get that certificated by similarity to the newer planes.
Military planes are a whole different animal; commercial aviation has barely introduced fly by wire, let alone flight by light. The industry is very conservative and the fact is that there hasn’t been much NEED to introduce video, although there are many arguments for it. It could help, but I’m not sure it really would have been a difference maker in any recent investigation, and so the reasons not to do it win out. The politics of it are weird, for sure.
Not really - that requires the introduction of such an antenna and data collection system (how would that be much different than storing it onboard?) as well as data centers that would hold thousands of hours of irrelevant and useless flight data from the thousands of flights happening worldwide every day.
The black boxes typically do survive; they are designed to do so. I think the only modern crashes where they were not recovered was 9/11, and the cause of those crashes were pretty well known. It might be very difficult to recover, like AF447, but even so, it was doable. Improving the 'pingers", ELTs, etc is a technologicaly easier and cheaper way of accomplishing a similar level of reliability.
A key issue with this, besides privacy and field of view and data storage etc., is the one of usefulness. Think about it, how useful really is a video from the cockpit ever going to be? What information could it ever provide that the flight data & cockpit voice recorders & ATC logs wouldn’t? In terms of air incursions (mid-air collisions & near-hits) well: A) Those are **incredibly **rare, and B) For jet aircraft they happen too fast for visual information to be of any use. If you’ve ever watched a recreation documentary you see that another plane goes from a dot, to a blur, to a CRASH! far faster than the human eye can resolve. Add in nighttime and cockpit video becomes doubly useless.
Another factor is everything in aircraft technology proceeds very cautiously, with many incremental steps along the way. New-fangled equipment is never adopted quickly (if at all). Video information gathering ***is ***increasing in aircraft, but very slowly. It will inevitably be ubiquitous, but it simply isn’t that useful a safety feature to demand fast, universal adoption.