An interesting discussion. UAS have been around for a long time - I was indirectly working with them in the early 1970's they are continually becoming more sophisticated. Almost all the UAS today are military or 'public' owned they are not commercial aircraft. This is not because commercial operators do not wish to have them; it is that regulations do not allow commercial operations. For this reason the accident rate is a little misleading as military operators, especially in war zones view their UAS to some extent as consumables accepting an attrition rate hugely higher than any commercial operator could countenance. This is the raison d'etre for UAS - they are sent into conditions that could not be risked by manned aircraft. See this brief article about Lockheed KMAX
Navy, Marines Share Lessons of Cargo UAV Missions in Afghanistan | Defense Update Portal. The KMAX like Boeing's Little Bird are 'Optionally Manned' aircraft; both can carry passengers (not necessarily pilots) this could be used for casevac operations in dangerous conditions.
Optionally piloted aircraft modifications of existing aircraft are not limited to slow aircraft see
Air Force Flies QF-16 Target Drone | Aero-News Network the automated Optional UAS version of the F-16.
Given that these UAS are being flown into conditions that would stretch the capabilities of an on-board pilot, it is difficult to continually claim that an on-board pilot is necessary when conditions are difficult. There is a huge difference between an automated pilotless aircraft and a remotely piloted aircraft. Currently the nascent certification systems are geared to remotely piloted aircraft not to automated pilotless aircraft. The pilots in command are expected to be required to have the same instrument ratings etc as the pilot of a manned aircraft. This reduces the attractiveness (from the accountant's point of view) for the idea of an unmanned airliner as the costs will be very similar any savings in manpower being swallowed by expected increases in hull insurance.
I fully expect that there will be moves toward reducing crews. In a way we are already seeing this with so called 'cruise pilots' - perhaps the argument will be made that only one 'cruise pilot' need be in the cockpit if a remote pilot can be alerted to problems and fly the aircraft. Remember, two man operation was inconceivable only a few years ago, yet now is standard practice. As automation becomes more and more intrusive with ACAS-X and automated TCAS RA response by the FMC; super density operations where the pilot is not allowed (OK extremely strongly discouraged) to take over from automation (as is the case now with RVSM airspace); RNP 0.1 approaches etc. etc.; It will become normal for pilots to rarely if ever control the aircraft. How many hands-on control minutes does the average pilot get in a transoceanic flight? - 5 minutes? It is a creeping move towards full automation.
I expect one pilot operation in freighter aircraft first with the 'safety pilot(s)' being remote and able to fly the entire trip if necessary.
Would I fly on a totally automated aircraft? With some of the pilots I have flown with I could even prefer full automation.
