Precise manual control of altitude happens all the time in civilian use. First example: tours. The reason helicopters are used is because they have no legal li

Precise manual control of altitude happens all the time in civilian use. First example: tours. The reason helicopters are used is because they have no legal limit how low they can go. Imagine booking a tour in a plane where you have to stay 500 feet above your target. No fun. But if you're going to be only 10 feet above your target (and in the case of the Grand Canyon tours, that can mean being below cliff faces), you had better have manual control of altitude. Second example: life flight or air ambulance. When there is an automatic way to control altitude like you are suggesting, the aircraft will probably only allow you to use one "profile" for landing. The profile dictates how steep your descent is. And it will usually not be very steep, but it will be a descent. But in life flight, you don't know your landing site. The only thing close by might be a meadow that requires a steep descent. Or a mountain face that requires a flat descent. Manual control of altitude is crucial to life flight. Third example: autorotations. They are the number one emergency maneuver. Basically, your engine goes out. What do you do in a helicopter when the engine drives your wings? The answer is that you let the rotor rotate freely, force air up through it (instead of it forcing air down) to keep it spinning, and then use that momentum close to the ground to slow your descent so that you can land softer than a baby's bottom. Having done that trick a few times, I know that you have to use collective. All of it. You have to use the other controls too, but those are to counteract the forces made by the collective, so they are just to keep you stable. (For example, when you activate the collective, it will bite into the air more, which will make the helicopter want to turn the opposite direction, so you put in some foot pedal.) So at the end of autorotation, you bring the collective all the way up; it bites into the air, and its momentum forces some air down, which lands you softly if you timed it right. The computer doesn't know how far above the ground you are, and more importantly, in an engine out situation, the computer has no power! Also, going into the autorotation requires manual control because you slam the collective down as fast as you can and put the helicopter into free fall! Yes, I'm not joking. You want air coming up through the blades immediately. Otherwise, they'll lose their momentum and fold upwards like in cartoons. And then you die. Yes, helicopter blades have to keep a certain RPM because centrifugal force is the major force keeping them straight. "RPM is life," my instructors said. I hope this answers your question. Personally, I would never step foot in such a helicopter without reading its full manual about what the computer does in emergency situations and how to handle an engine out/computer out emergency. And I'd want to look at the software itself (since I'm a programmer). The thing with aviation is that autopilots can do everything except emergencies. Can you imagine an autopilot performing the Miracle on the Hudson? I can't. That takes raw skill and intelligence. That's why there are still two pilots up front when the whole flight is so automated that they may not even touch the controls. And the FAA will always require them up there. Pro tip: anytime a new aviation tech comes out, you should think, "Will this help in emergencies, hinder, or be neutral?" As long as it doesn't hinder, it may get used. A new type of engine? As long as it's just as reliable, it's neutral in emergencies. The Cirrus Airframe Parachute System? Helps in emergencies. Pilots love it. This? This will only take options away from a pilot during emergencies. Airbus will struggle to get it adopted.