Fly By Wire: Possible "early warnings"
Sorry for the very long post, but I think it might be interesting for some.
There are two advantages of Fly By Wire systems, the first is obvious and very important on Military aircraft: some aircraft are inherently unstable and with Fly By Wire systems thery are still flyable, which is a huge bonus. In Civil aviation, FBW systems are here to help the pilots. Make their jobs easier and safer. Fly By Wire system is good, because the aircraft understands what do you want to do, and can also let you know what is safe to do.
Either by some restrictions in normal law, or with any feedback the system provides. One of the risks for Fly By Wire system is simple: If we assume, "if it is ok by the FBW system, it should be safe", yet most current FBW implementations doesn't monitor the consequences of our actions.
The FBW system knows any mach speed that is greater than stall speed and lower than overspeed is safe. There can be some margin, etc. but most current FBW systems doesn't try to predict possible outcomes and determine "how safe this action is" based on that. Simply because when these systems were designed a complex software that would try to judge multiple parameters in context would require very strong hardware, and a lot of research.
Any precise enough simultion of possible consequences would be too complex to be feasible if we consider what kind of computer hardware was present, and what could be installed on an Aircraft.
In fact the Fly By Wire software doesn't even recognize, that if it switches to alternate law, or direct law while the pilot is trying to do something, the results can be unexpected. (If the system doesn't limit the pilot but only provides feedback, it isn't important) And you can't depend on fly by wire system to know what is safe / what is unsafe.
Lets go back to mach speed.
If you can correct if you get close to a stall and can descent safely, your speed can be safe, even if it is close to stall speed. If you expect the wind is constant and nothing will happen: who cares.
If wind isn't constant, so your airspeed can be far from constant (thanks to turbulence), and there is something under you (either terrain, or a CB, etc) your speed might be far less safe.
Some speed can be safe in most cases, but if you can expect turbulence (thanks to weather) it can be too close to both overspeed and stall speed, etc. and can be risky.
Yet, the Fly By Wire system won't even offer any feedback about possible hazards from maintaining that speed and altitude. If the pilot assumes, it is safe (after all FBW software considers it safe), that can lead to nasty incidents.
If there is a risk of icing (and loss of airspeed data) you might prefer if the difference between stall and overspeed is significant, but if it isn't you won't get any warning.
If you trust "What you do is safe, because the FBW aircraft lets you do it" sooner or later you will run into problems.
I heard some criticism about Airbus aircraft, saying many pilots are simply "operating the system" and not flying the aircraft in many cases, and depend too much on the system.
Since there can be any software problem, or any problem with the instruments, I prefer if the pilot has the final say, and you don't have to disable many other instruments to have access to alternate law / direct law. So I prefer feedback over restrictions. (If FBW system gets incorrent airspeed data, but doesn't detect the fault, and restricts the choices of pilots based on it, that can be risky)
But if we are using FBW and related system, I would prefer if it would know "how safe this speed / maneuver, etc is in current circustances" in addition to "this is within some predetermined limits". And it would provide feedback based on it.
And if it restricts the pilot, it would warn about switching to alternate law, so it doesn't surprise the pilots.
While I am not sure if any such warnings could have helped to avoid the AF447 accident, I think early warnings that are shown before things get rough can be useful for preventing future problems.
Monitoring more data from instruments in real time, and predicting possible problems can be easier with modern computer systems. And to my best knowledge the required research is minimal (thanks to UAVs capable of both autonomous flight and remote controlled operation).
What do you think, would such early warnings be useful in civil aviation?