It seems strange that trucks and taxis can be set up to transmit GPS data in real time but aircraft cannot. Wouldn't this be a useful function and feasible to implement?
Most commercial aircraft transmit their GPS-based position twice per second. This is part of their Automatic Dependent Surveillance – Broadcast (ADS–B) broadcasts. The problem with providing world-wide receiver coverage for this system is that the frequency it uses only travels line via line of sight, so it won't travel past the horizon.
Providing coverage over large bodies of water would require a network of buoys, which would be quite expensive.
Another possibility is to put ADS-B receivers on satellites. This concept has been developed by (Aireon) on the Iridium Next satellite constellation. The system (66 satellites + 9 spares) became operational in 2019 and provides global coverage.
The following is, in respect of aircraft, speculation. However, I did spend a decade working on the GPS based tracking systems mentioned in the question, with particular emphasis on remote operations in rural Australia. We did trial the use of Iridium, so I will comment on that too.
Firstly such systems do not report in real time, that is baloney put out by marketroids who don't even know what real-time means (I discovered this while trying to persuade them not to put unsupportable claims in promotional material.)
Vehicle positioning and telemetry systems report periodically. The more sophisticated units are store and forward, which means they log position at frequent intervals and upload bursts of telemetry as and when the network is available. The "real-time" ones use the telephone network, while others upload as they pass through WIFI waypoints, usually at depots and pickup locations. Hybrids exist, reporting sparse data periodically through the network and uploading higher resolution sampling at waypoints. The more sophisticated units have rules for priority.
Land freight in the United States runs along narrow concentrated corridors (roads and rails) that are well serviced by the telephone network. This is also true for freight on the eastern seaboard of Australia, but availability diminishes as you head west. As a result, the network is available most of the time so the units can upload at regular intervals, maintaining the illusion of real time tracking. Some systems also support manual polling of unit to obtain fresher position information on demand but this only works when the unit happens to be on the network.
When we tried it, Iridium had dismal bandwidth at outrageous cost. In the southern hemisphere, the satellite density is so low that you might see a satellite briefly every twenty minutes or so. The time between deciding things are out of control and the aircraft crashing could easily see the plane down before a satellite connection is available, although I suppose things might be better at high altitude with much wider horizons. The Iridium systems I had were power greedy, but that's not a problem in vehicle mounted systems.
Aircraft move too fast for the cellphone network; by the time they finish negotiating with one cell they're over the next. Also their altitude means they get similar signal strength from several neighbouring cells making them hop between cells unpredictably. I have been in a private plane in which the radio equipment failed, and the pilot had a devil of a time keeping a cellphone connection to the tower due to speed and altitude. Commercial craft are higher and faster so I doubt it would work at all.
Out at sea, there's just nothing but satellite comms. Over the Atlantic I imagine it wouldn't be too bad, but there are a lot of commercial routes that go a long way in the middle of big wet nowhere. Also there are a lot of commercial flights, and they move so fast you'd have to report every second to get any kind of positional precision. I'm not sure the civilian satellite network could cope. I know they stream vast amounts of TV data, but (a) that bandwidth is committed, and (b) that's loss tolerant streaming. It's high lag. When you try to use them to support non-loss network chatter the throughput plummets.
I have no idea why the black box isn't ejected prior to impact. You could easily set up a system that once armed would eject it when the craft dropped below a certain altitude. If it were me there'd be a little parachute and a self-inflating weather balloon, and the plane would be transmitting position updates to the ejected system until it failed. A little solar charger could keep the transponder going and with the balloon keeping the unit up off the water it shouldn't be hard to find. Such a system would have no network dependencies and although it might well drift a long way before being collected, it would know where it was at the time of ejection and exactly where the aircraft was when it ceased transmitting.
An interesting question from the comments:
Perhaps the aircraft could maintain an adaptive network between them as they fly around? That would increase the range of the coverage. The bandwidth for reasonably periodic position packets is low enough that one aircraft could easily forward packets from dozens of others, (and then on to base stations).
It's a nice idea but oceans are colossal. Even with hundreds of flights in the air at the same time, even in air corridors over land, being in visual range is a periodic event. Only in the crowded airspace around a commercial airport is this likely to work - and when you're that close you have comms anyway. A shame because it's creative and technically interesting.
The main problem with this is that it is somewhat easy to know where the device is, it's a lot harder for the device to tell someone else where it is. Satellites are really the only option for this sort of thing, and a lot of the polar routes have pretty poor satellite coverage anyways. Also, satellite bandwidth is quite expensive.
there is already something like that, the primary and secondary radars allows for positional data to be received by the controller, so adding this is superfluous from FAA's standpoint. To do this voluntarily is expensive.
To roll out a transmission of gps position data you need someone to be listening to it. Cars use the existing mobile network. Planes would likely use the transponder frequencies.
You also still need to be compatible to existing transponders while the global rollout happens.
From a documentary on the 2009 Air France crash: I remember an aviation expert answering the same question. He said that it is expensive (About extra $300 million for a typical national carrier per year) and Airliners are reluctant to take on any additional expenses even if such an undertaking improves safety (The Air France jet crashed because the Pitot tubes froze thus disengaging the auto pilot. The Airline was aware that the Pitot tubes needed to be replaced way before the accident but did nothing because it was expensive).
No, it wouldn't be useful. And no, it would be exceedingly hard to implement, not to mention exceedingly expensive.
Cars and trucks are rarely out of range of cellphone towers, which is the system they use to transmit that data to their company headquarters. Mind that that's only used for some trucking and taxi companies, and a few car rental and lease companies, not generally (though there are some governments that are pushing for it, which has very strong privacy considerations as essentially your location would be known to the government within a short radius at any time, Big Brother is watching you...).
It's used mainly for fleet management purposes, like a rental agency can use it to pinpoint you to being where that speeding ticket was written so they can tell police who to send the fine to (and the points on your license).
The equipment for cars and trucks is also small, weighing in at maybe a pound (including mounting brackets, cables, etc.) and drawing little power.
For aircraft, it would need to be larger as instead of a small cellphone transceiver it would need a satellite radio link and antenna. It would also as a result draw a lot more power, which means higher fuel use on top of the extra fuel use from the higher weight of the aircraft (ignoring the possible extra drag from the antenna which makes it even worse).
And it's not real time, obviously, the cost would get to be far too high. Most cars and trucks would send an update only once every few minutes at most (selectable).
And aircraft already have something like it anyway, with systems that can send location and status data to the airline. But it's triggered manually, usually, and not in universal use.
Such a system would have virtually no purpose, and a high cost of operating it. You'd need not just a small box in every aircraft, you'd need a network of communications satellites, ground stations around the world with 24/7 staff on hand, etc. etc.
And I seriously doubt you're going to want to add a few hundred dollars to the price of every ticket just in the off chance that when your flight goes down someone can find the wreck a bit sooner.
Actually I don't think it's terribly difficult to implement technically. Of course getting regulators and airlines on board - and dealing with security and privacy issues are other matters entirely.
As others have pointed out there are existing satellite data networks (e.g. Iridium and Inmarsat) that can be used to transmit location information "back to base". The field terminal equipment for these networks isn't particularly power hungry either.
Infact a company I did a bit of work with in the past (on another project) sells a system exactly like that right now, which is designed for helicopters and small aircraft. They're not a big company, so I'd lay odds that they have competitors in this field too.
https://www.indigosat.com has details of their particular system.
I haven't read through all of these responses, but most focus on locating the plane going down in the middle of the ocean, as well as major new equipment and transmitters. First, the location wouldn't have to be a separate GPS transmitter. It only needs to be voice or text transmitting the location data which the plane already has. Whatever the communication of the location data, it doesn't have to be relayed at the point of impact or damage. Obviously, that would be most helpful, but in the case of MH370 - and especially if it was flying in a straight line, then having 2 points would be helpful.
That is, if the plane is detected over Malaysia, then over the Indian Ocean, they would have a path and relative search area based on fuel, etc. It's not perfect, but simply a transponder that can't be turned off would probably have been sufficient enough to detect the general area of the aircraft in the first week. There needs to be satellite tracking in real-time, but I think there are sufficient systems in place already that would have made this much easier.
As an example, and in response to the earlier post about \$300M system, if a person sat on a plane and texted lat/long every 15 seconds, they would find this plane much quicker even if it went beyond text towers. I'm not suggesting having a person do that, but you can create a device that does that for a relatively small amount of \$. It's not the perfect system, but is better than nothing.
This service is already available for GA aircraft on VFR flight plans, using the consumer grade Spot Beacon. (Link to Press Release).
Because Spot covers most of the earth, I cannot see why it wouldn't be a generalized solution.
The cost (at the GA level) is around \$100 for the device and about \$100 per year, so its quite reasonable.
The device is not covered by a TSO, so it is not certified for commercial use, and such a certification would be pricey and difficult. Still, I think the technology is ready and has been demonstrated.
Some 400 aircraft already do this via the AFIRS system, which uses Iridium satellite constellations to communicate. It costs some $100K to install.
Note that Inmarsat has a free tracking offer, free as in customers that are paying for services from third-party providers. Doesn't have complete worldwide coverage.
As to why it hasn't been mandated, I do believe it has to do with necessity with cost benefit analysis / risk management vectors until recently not requiring it. With the search costs of AF447 reaching 100M and the at least $80M spent on the MH370 search, the risk profile may now provide some impetus to install such technology. If only to reduce cost exposure in the event of a loss and to gain the advantages provided by real-time monitoring of aircraft systems.
Bottom line is that it hasn't been seen as a necessity, but it is technological a reality already with some aircraft already carrying the systems.
As stated previously bandwidth cost and coverage of satellite communication accross the ocean is a major issue for something which doesn't directly make the airlines make money (contrary to in flight data for PAX). Even if you do it too often you could reach the limits of the actual satellite communication system.
This is well studied by the report made following the AF447 crash : http://www.bea.aero/en/enquetes/flight.af.447/triggered.transmission.of.flight.data.pdf
To reduce the cost of it the question asked is "can the plane transfer its position when it detects an unusual situation".
Consider that radio line-of-sight horizon varies with altitude and transmission power. I have not calculated the radio line-of-sight horizon from 30,000 to 35,000 feet but I imagine that it is staggering.
There are Ham radio operators listening for distress calls twenty-four hours a day, around the world. A distress call accompanied by GPS coordinates would get immediate attention from anyone on the frequency. Ham satellites are another resource.