PPRuNe Forums - View Single Post - Indonesian aircraft missing off Jakarta
View Single Post
Old 4th Nov 2018, 16:57
  #573 (permalink)  
Ian W
 
Join Date: Dec 2006
Location: Florida and wherever my laptop is
Posts: 1,350
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by A Squared
Well, one of the lines of speculation is that the handling problems were exacerbated by the computer control of an automated trim system, as a result of erroneous data being fed into the computer. What you're proposing is to have that same computer (whcih is making the problem worse, based on bad data input), take over flying the airplane, and fix the problems that it is creating ... based on that same bad data that is causing the problem. See how that all gets kinda circular?
'The computer' (possibly several all working together in a network) already knows that there is a problem with the pressure instruments as the symptoms are reasonably obvious and can be diagnosed 'by the computer' with no problem, except that nobody has asked the software designer/architect/analyst to do that. The reliance has always been on handing the bag of bolts to the crew with a cryptic message or three together with aural alerts that might assist or confuse.
A system could easily alert that GPS alt changes mismatch pressure altitude change and offer to provide GPS altitude as a reference an on altimeter. Yes this does not match pressure altitude but is more accurate measuring vertical distance and rate of climb and descent. The much derided magenta line is generated by 'the computer' which also knows it in 4 dimensions how fast the aircraft is traveling along it and whether the aircraft is above or below it. 'The computer' in short could disregard the pressure instruments and just follow the magenta line in all 4 dimensions. It could even generate a new magenta line to the appropriate runway threshold and fly a recovery. Yes it may be hunting a little but it will know the approximate engine settings and then work from that reference. 'The computer' would then be doing the flying the aircraft by reference to the surface in the way that a pilot could do if that pilot stops trying to make sense of unreliable instruments and just looks out of the window disregards the instruments and flies by setting pitch and power. (A repeated mantra during the AF447 thread)

Another way of putting that is first thing to do is aviate - not try to solve system problems. 'The computer' could be designed to do that too. Perhaps it should be.
Ian W is offline