4 decades ago, someone who seemed to know about helicopters said that hovering in one was like balancing on a ball (like bears do in a circus). But it seems like by using computers hovering and other helicopter control could be more or less automated - is this correct?
Yes, a lot of helicopters are easier to fly today due to automation.
Modern (higher-price) helicopters are equipped with autpilot modules which typically range from pilot assistance to full-fledged four-axis autopilot with function from automatically following airspeed, heading and height setpoints over autonomous waypoint flying to features like automatic hover control. A good example is the Helionix suite from Airbus Helicopters which offers exactly these features. Other examples are for example the Bell 505, the Bell 525 (which even features full fly-by-wire) or modern military helicopters like the Eurocopter Tiger, the CH-53K (also fly-by-wire) or the NH-90 (which was the first production helicopter featuring a full-fly-by-wire system).
Perhaps a word on how control is achieved without having a full fly-by-wire system at the example of the Airbus Helicopters (which I happen to know a bit): This control is achieved by inserting actuators with limited bandwidth and control authority (therefore with limited travel) into the control linkage with the addition of a trim actuator. The setup is nicely described here SEMA (smart electro-mechanical actuators). This limited control is then used by the autopilot to provide all its functions. You can also see those Actuators (marked in red as AFCS Actuators) in this picture of the control linkages of a Sikorsky S76:
If the helicopter is flown manually, the Stability augmentation System (SAS) is turned on by standard, which is already a great help. Often you can also choose to select a higher control mode such as the Attitude Command Attitude Hold (ACAH) mode, which lets you directly command the roll/pitch attitude of the helicopter while the autopilot takes care of fine control.
With these systems, helicopters are a lot easier to fly. From personal experience I can tell you that even untrained people have no problem to fly and hover helicopters equipped with such autopilots. However if the autopilot augmentation is switched off, the helicopter is back to "balancing on a ball" just like it was 40 years ago due to the inherent unstable dynamics of helicopters!
So yes, with the right (and expensive) helicopter, it is easier then it was 4 decades ago to fly helicopters.
My work on helicopter flight controls started more than 30 years ago, and at that time there were already helicopters is service with Automatic Flight Control Systems (AFCS) and Stability Augmentation Systems (SAS). And computers were very expensive at the time.
Fixed wing aeroplanes are designed so that the airframe is inherently stable: when push comes to shove and all systems are down, the pilot is able to maintain control over a stable platform. Exceptions are to be found in the military field, one of the first being the F-16 which had fail-operational inner loop flight control computers (as mentioned in this answer.
Helicopters cannot be designed such that the airframe is always inherently stable - that is, when the rotor is on top.. Helicopters are:
- Unstable in the hover, indeed like balancing on top of a large ball. The instability eigenfrequencies have a large enough time constant for humans to learn how to maintain equilibrium.
- Subject to cross couplings of the flight controls, which is why larger and more expensive helicopters have mechanical mixers as well.
So automated helicopter SAS has been around for those 4 decades - in more expensive, larger helicopters. Where the SAS is not flight critical and when hovering without it those balancing skills still need to be provided by the pilot.
Complete and always functional SAS requires the system to be fail-operational, by providing redundant system functionality. The F-16 has four identical systems, upon one system fail the other three vote out the failed system, when a second one fails the remaining two can still vote out the failed one, the third failure makes life hard because the airframe is unstable. It is these considerations that make flight critical automated systems expensive, not the cost of the processors and sensors.
Pic from the wiki
Fail-passive SAS in a Robinson R-22 might become a feature in the future, but will the market really require this feature in a low-cost helicopter that is used for teaching unstable hovering skills?