4 ms·

A relatively simple fix that could prevent planes from disappearing

29 points by joshbaptiste 12y ago

joshbaptiste 12y ago
I love the idea this article proposes, basically Waze for airplanes where real time weather data comes from the actual flights themselves and shared via traffic controllers or the planes themselves.
eveningcoffee 12y ago
I am a bit surprised that he did not mention big data. But there is point in this talk. Of course this system will not cost 300 as it has to be dependable and it must kept properly calibrated.
TeMPOraL 12y ago
Can someone explain to me why they can't just stick a GPS receiver wired to Iridium satellite phone and duct tape it to the back of the plane? I.e. it shouldn't be neither hard nor that expensive to send a few dozen bytes of telemetry every minute via Iridium. Why anything like this is not happening?
- jpollock 12y ago
  They charge a lot of money for real-time telemetry from a plane. This charge can frequently represent the entire profit for a flight for smaller carriers, so they don't elect to purchase it.
  - rational-future 12y ago
    Who is 'they'? You don't even need Iridium, a shortwave radio transmitter will suffice.
    - achamayou 12y ago
      Boeing or Airbus. Look up ACARS, it exists, many planes are equipped. I find conflicting information about the Air Asia flight, a Bloomberg article claims the plane wasn't equipped, other places claim it was, but the airline wasn't subscribing to the data it generated. It's not clear at all whether this is related to the cost of the system itself however.
    - spacefight 12y ago
      Would it? Could you elaborate?
  - ars 12y ago
    You don't need real-time (a small delay is OK) or much telemetry. Just 10 bytes or so every 5-10 minutes. And over land you can use cell phone tower, so smaller carriers (who don't go much over water) would not be impacted badly. Actually over land you probably don't need anything at all, so this really would just impact those best able to handle the expense.
- deanclatworthy 12y ago
  I think this is the question everyone is wondering. But even in a minute a plane travels a huge distance. And also keep in mind the height at which they fly so assuming the planes don't stop going forward they can travel huge distances on their descent to the ground. Nevertheless, I also wonder why this couldn't work and just send the data more frequently. I don't know anything about the GPS consumers connect to but if my TomTom can connect to a satellite in space and give me my location down to a few metres at a regular interval then why can't a plane do the same?
  - SapphireSun 12y ago
    For what it's worth, GPS works because the satellites are broadcasting signals to earth, not your phone to space.
  - TeMPOraL 12y ago
    > But even in a minute a plane travels a huge distance. And also keep in mind the height at which they fly so assuming the planes don't stop going forward they can travel huge distances on their descent to the ground. Even a few dozen bytes worth of data every minute about current position, heading, airspeed, heading and speed of the wind is enough to reduce this theoretical "huge distance" (that is much smaller than what people have to deal with without tracking data) to a relatively small cone that can be searched effectively.
- toomuchtodo 12y ago
  They are. New Iridium birds are going up with ADS-B receivers onboard, with the data relayed to the ground over the Iridium network. http://www.aireon.com/AboutAireon/AnIridiumInnovation http://www.aireon.com/AboutAireon/AnIridiumInnovation Replacing orbiting equipment takes time and money.
tokenadult 12y ago
It's important to note that the substance of the article here is not about improving tracking of airplanes while they are flying (which actually does very little to keep them from disappearing if a catastrophe stops them from flying) but is actually about tracking weather more accurately by gaining real-time data from other aircraft in the vicinity. This suggestion might indeed have helped prevent the most recent Air Asia crash, and perhaps also the Air France crash from several years ago. It takes good information to respond to rapidly developing storms at high altitude over the ocean, and to advise pilots how to change their courses (along heavily trafficked airline routes) to avoid both dangerous weather and collisions with other planes.
- tacticus 12y ago
  How exactly would it help af447? wasn't that determined to be an issue with the pitot tube becoming blocked and one of the pilots just holding the stick back until they stalled at 47 thousand feet.
  - 69_years_and 12y ago
    The fact that there was a storm that could cause icing (perhaps due to super cooled water, or is it super chilled water) could have been shared thus giving AF447 a greater opportunity to avoid it, which could/would have meant the pitot's would not have become blocked which would mean the auto pilot would not have turned off due to faulty airspeed input which would mean there would have been no need for pilot to fly the plane and hold the stick back for so long. Of course its a moot point for AF447 - RIP.
upofadown 12y ago
The Microsoft project is a way of figuring out current upper wind conditions from data available from the aircraft that fly at those upper altitudes ... which is pretty cool, but doesn't help plan fuel loads by itself. For that you would have to use those derived wind speeds to generate a better forecast. You need to know what the winds are going to do in the future. Getting the upper wind forecast wrong doesn't cause a crash. At the most it might cause a diversion. The Microsoft idea has nothing to do with the sort of bad weather that people think might of been an issue with Flight QZ8501. That bad weather mostly consists of thunderstorms which are small unpredictable areas of vertical air movement. Drones are not likely to be of much help here as they normally operate at low altitudes.
fubarred 12y ago
The most terrible reality is that the capability for tracking virtually all long-haul airframes (99.999%) already exists. It's called Inmarsat FANS-1/a ADS-C, which is now being offered to all airlines for free, post-MH370. It was present on the MH370 airframe (but for a cost), and could've located it to within tens of square miles instead of millions. http://www.inmarsat.com/news/inmarsat-provide-free-global-airline-tracking-service/ http://www.inmarsat.com/news/inmarsat-provide-free-global-ai... http://www51.honeywell.com/aero/common/documents/FMS_DirectTo_3rd_Editionrev1.pdf.pdf http://www51.honeywell.com/aero/common/documents/FMS_DirectT... http://members.optusnet.com.au/~cjr/CPDLC.htm http://members.optusnet.com.au/~cjr/CPDLC.htm
alblue 12y ago
This article really demonstrates the level of idiocy that armchair experts contribute to the discussion. It's as if the author had to publish a piece on the disappearance and use it to generate as many click bait responses as possible. Firstly, for ground covered aircraft there is already extensive weather radar (the same used to demonstrate clouds on your tv's weather graphics on the evening news) which shows what is happening live. This is tracked by most airline operators and the hotspots can be rerouted from central if necessary. Secondly most newer aircraft have a forward looking weather radar that can warn them as the plane flies of the immediate vicinity (though they don't have full visibility of what's on the other side). At the very least they can use this to do what happened here; request a variation in the path of the flight. Thirdly the main problem of weather is in areas not covered by weather radar. This is where drones would be most useful; but that is also the problem: they need sufficient fuel to get them to stay in or around one place and unless there is automated return to base and refuelling options it isn't going to happen. Fourthly the planet is big. Really big. You may think it's a long way down to the chemist at the end of the street etc. the point is knowing what the weather is like tens of miles away isn't really that useful for identifying the kind of cells causing bad weather which are generally known about in any case. To get the kind of required resolution you either have to have lots and lots of them, or send a leading drone out in front of the aircraft's intended path waving a red flag. Except to do that it has to fly at a similar speed to the jet which means each drone is going to cost many millions In hardware and fuel for each flight. Lastly no pilot is going to want to fly through a mid air drone field because the chance of disaster occurring due to high speed impact with a stationary object is much higher than that of suffering weather damage. The reason planes fly at different altitudes depending on which direction they are travelling is to avoid meeting traffic coming the other way; generally because if you notice it it's already too late to avoid it.

29 points by joshbaptiste 12y ago

I love the idea this article proposes, basically Waze for airplanes where real time weather data comes from the actual flights themselves and shared via traffic controllers or the planes themselves.

eveningcoffee 12y ago

I am a bit surprised that he did not mention big data. But there is point in this talk. Of course this system will not cost 300 as it has to be dependable and it must kept properly calibrated.

TeMPOraL 12y ago

Can someone explain to me why they can't just stick a GPS receiver wired to Iridium satellite phone and duct tape it to the back of the plane? I.e. it shouldn't be neither hard nor that expensive to send a few dozen bytes of telemetry every minute via Iridium. Why anything like this is not happening?

jpollock 12y ago
They charge a lot of money for real-time telemetry from a plane. This charge can frequently represent the entire profit for a flight for smaller carriers, so they don't elect to purchase it.
- rational-future 12y ago
  Who is 'they'? You don't even need Iridium, a shortwave radio transmitter will suffice.
  - achamayou 12y ago
    Boeing or Airbus. Look up ACARS, it exists, many planes are equipped. I find conflicting information about the Air Asia flight, a Bloomberg article claims the plane wasn't equipped, other places claim it was, but the airline wasn't subscribing to the data it generated. It's not clear at all whether this is related to the cost of the system itself however.
  - spacefight 12y ago
    Would it? Could you elaborate?
- ars 12y ago
  You don't need real-time (a small delay is OK) or much telemetry. Just 10 bytes or so every 5-10 minutes. And over land you can use cell phone tower, so smaller carriers (who don't go much over water) would not be impacted badly. Actually over land you probably don't need anything at all, so this really would just impact those best able to handle the expense.
deanclatworthy 12y ago
I think this is the question everyone is wondering. But even in a minute a plane travels a huge distance. And also keep in mind the height at which they fly so assuming the planes don't stop going forward they can travel huge distances on their descent to the ground. Nevertheless, I also wonder why this couldn't work and just send the data more frequently. I don't know anything about the GPS consumers connect to but if my TomTom can connect to a satellite in space and give me my location down to a few metres at a regular interval then why can't a plane do the same?
- SapphireSun 12y ago
  For what it's worth, GPS works because the satellites are broadcasting signals to earth, not your phone to space.
- TeMPOraL 12y ago
  > But even in a minute a plane travels a huge distance. And also keep in mind the height at which they fly so assuming the planes don't stop going forward they can travel huge distances on their descent to the ground. Even a few dozen bytes worth of data every minute about current position, heading, airspeed, heading and speed of the wind is enough to reduce this theoretical "huge distance" (that is much smaller than what people have to deal with without tracking data) to a relatively small cone that can be searched effectively.
toomuchtodo 12y ago
They are. New Iridium birds are going up with ADS-B receivers onboard, with the data relayed to the ground over the Iridium network. http://www.aireon.com/AboutAireon/AnIridiumInnovation http://www.aireon.com/AboutAireon/AnIridiumInnovation Replacing orbiting equipment takes time and money.

tokenadult 12y ago

It's important to note that the substance of the article here is not about improving tracking of airplanes while they are flying (which actually does very little to keep them from disappearing if a catastrophe stops them from flying) but is actually about tracking weather more accurately by gaining real-time data from other aircraft in the vicinity. This suggestion might indeed have helped prevent the most recent Air Asia crash, and perhaps also the Air France crash from several years ago. It takes good information to respond to rapidly developing storms at high altitude over the ocean, and to advise pilots how to change their courses (along heavily trafficked airline routes) to avoid both dangerous weather and collisions with other planes.

tacticus 12y ago
How exactly would it help af447? wasn't that determined to be an issue with the pitot tube becoming blocked and one of the pilots just holding the stick back until they stalled at 47 thousand feet.
- 69_years_and 12y ago
  The fact that there was a storm that could cause icing (perhaps due to super cooled water, or is it super chilled water) could have been shared thus giving AF447 a greater opportunity to avoid it, which could/would have meant the pitot's would not have become blocked which would mean the auto pilot would not have turned off due to faulty airspeed input which would mean there would have been no need for pilot to fly the plane and hold the stick back for so long. Of course its a moot point for AF447 - RIP.

upofadown 12y ago

The Microsoft project is a way of figuring out current upper wind conditions from data available from the aircraft that fly at those upper altitudes ... which is pretty cool, but doesn't help plan fuel loads by itself. For that you would have to use those derived wind speeds to generate a better forecast. You need to know what the winds are going to do in the future. Getting the upper wind forecast wrong doesn't cause a crash. At the most it might cause a diversion. The Microsoft idea has nothing to do with the sort of bad weather that people think might of been an issue with Flight QZ8501. That bad weather mostly consists of thunderstorms which are small unpredictable areas of vertical air movement. Drones are not likely to be of much help here as they normally operate at low altitudes.

fubarred 12y ago

The most terrible reality is that the capability for tracking virtually all long-haul airframes (99.999%) already exists. It's called Inmarsat FANS-1/a ADS-C, which is now being offered to all airlines for free, post-MH370. It was present on the MH370 airframe (but for a cost), and could've located it to within tens of square miles instead of millions. http://www.inmarsat.com/news/inmarsat-provide-free-global-airline-tracking-service/ http://www.inmarsat.com/news/inmarsat-provide-free-global-ai... http://www51.honeywell.com/aero/common/documents/FMS_DirectTo_3rd_Editionrev1.pdf.pdf http://www51.honeywell.com/aero/common/documents/FMS_DirectT... http://members.optusnet.com.au/~cjr/CPDLC.htm http://members.optusnet.com.au/~cjr/CPDLC.htm

alblue 12y ago

This article really demonstrates the level of idiocy that armchair experts contribute to the discussion. It's as if the author had to publish a piece on the disappearance and use it to generate as many click bait responses as possible. Firstly, for ground covered aircraft there is already extensive weather radar (the same used to demonstrate clouds on your tv's weather graphics on the evening news) which shows what is happening live. This is tracked by most airline operators and the hotspots can be rerouted from central if necessary. Secondly most newer aircraft have a forward looking weather radar that can warn them as the plane flies of the immediate vicinity (though they don't have full visibility of what's on the other side). At the very least they can use this to do what happened here; request a variation in the path of the flight. Thirdly the main problem of weather is in areas not covered by weather radar. This is where drones would be most useful; but that is also the problem: they need sufficient fuel to get them to stay in or around one place and unless there is automated return to base and refuelling options it isn't going to happen. Fourthly the planet is big. Really big. You may think it's a long way down to the chemist at the end of the street etc. the point is knowing what the weather is like tens of miles away isn't really that useful for identifying the kind of cells causing bad weather which are generally known about in any case. To get the kind of required resolution you either have to have lots and lots of them, or send a leading drone out in front of the aircraft's intended path waving a red flag. Except to do that it has to fly at a similar speed to the jet which means each drone is going to cost many millions In hardware and fuel for each flight. Lastly no pilot is going to want to fly through a mid air drone field because the chance of disaster occurring due to high speed impact with a stationary object is much higher than that of suffering weather damage. The reason planes fly at different altitudes depending on which direction they are travelling is to avoid meeting traffic coming the other way; generally because if you notice it it's already too late to avoid it.