Date: 07 Oct 98 02:49:21 From: JF Mezei <email@example.com> References: 1 2 Followups: 1
View raw article or MIME structure
PS2727 wrote: > I agree that the maneuver should not have been done in that setting, but > my point is simply this- The Airbus approach to automation does not, in > my opinion, merge well with the way transport pilots have operated > aircraft. Strange as it may seem, there are many situations in airline > flying where crews revert to stick and rudder skills to save a rapidly > deteriorating situation. Assuming that bringing the nose down was actually the best way to get the aircraft to gain altitude and that in that accident, the computer was actually trying to get aircraft to gain altitude. What is wrong then with the pilot telling the aircraft "I want to go up" and then the aircraft calculating that in the current situation, the best way to go up is to actually bring the node down ? Or pilot asking the aircraft to turn left and the computer calculating that the best way to acheive this is to reduce thrust on one engine, work the rudder, elevators and ailerons. Obviously, it is a question of trust. That pilot probably did not trust the computer and thought that the computer was going nuts by doing the opposite of what it wanted it to do. But if the pilot were aware of what sort of logic it uses to calculate the best way out of a situation and the pilot realised that the computer was actually looking at more parameters than he could look at and hence calculating a better solution, perhaps the pilot would learn to be more comfortable with the system and learn to trust it.