This article shares some of the theme of a thread that is raging on beechtalk.com. We are discussing the coming end of the commercial pilot. The war in Iraq and Afghanistan are pushing the development of robots to the forefront. They are in the air, on the ground, and are being developed for deployment in the sea. My contention is that it is not a far stretch from the current UAVs to no human pilots in airliners. Sure, the technology and reliability have a long way to go. But there is a ton of money pushing these efforts. A fore taste of things to come is happening at FedEx. They are demonstrating a single pilot DC-10 where the pilot does nothing but monitor the airplane while it is controlled from the ground by a guy in a trailer. What is really interesting is the total denial evident on Beechtalk.com.

I could see automated vehicles under certain cercumstances, but I think when people are involved, there will be at least one pilot incase something goes wrong. As the automation becomes more advanced, the pilot will pretty much sit there and watch the instrumentation, but that will be about it. Of course, I'm only thinking about 50 years down the line, who knows how people will feel about automation in 100 years or greater.

Karl, no offense intended, but your response is so typical of those afraid of changing technology. They push it so far into the future to be of no consequence to them. At the time the first internal combustion engine was developed, did anyone have any idea it would turn into a transportation revolution? Heck, the oil industry was in its infancy too. When Edison was using his brain trust to develop the light bulb, he was hell bent on DC power. When Tesla invented AC power, Edison did everything he could to bury him.

It is interesting how history has treated Tesla. He invented an entire industry with only a small cadre of people. He was an odd fellow who only wanted to be left alone in his laboratory. If Westinghouse had not invested in his enterprise, we would be in the electrical dark ages now. On the other hand, Edison was a businessman and a consummate promoter. He left the inventing up to his legion of hired engineers.

Just look at the history of personal computers, especially the early years. Do you think IBM felt threatened at all?

Actually, I agree with Karl. Most (but not all) passenger rail has a human operator on-board the actual train and we've got the technology to automate this immediately. I don't think we'll see autonomous aircraft anytime soon, but single-pilot airlines - maybe.

One key requirement for highly automated flight is a good "detect and avoid" capability that works as well as, or better than, a pilot's Mark 2 eyeballs. A few companies were working on it a couple of years ago, but I've lost track of them. Has anyone developed the capability yet? Does anyone know who is working on the problem now?

It's quite possible that if the commercial/military/security benefits are thought to be strong enough, the rules will be re-written to make 'see and avoid' unnecessary for autonomous aircraft. Trains (which could be fully automated now) don't use 'see and avoid' - they restrict access to the tracks. It could be that autonomous aviation will cause IFR-only corridors to appear. This is likely to be done by a creeping extension of existing airspace limitations. Only GA would object to this, and it's pretty clear that GA has very limited political influence (especially when changes are for reasons of 'National Security').

IFR corridors for UAVs will not relieve an autonomous UAV from the Detect & Avoid requirement. The FAA has always required aircraft under IFR control to maintain visual separation when visibility permits. There is nothing to indicate that this policy will change. By the way, there are several thousand vehicle-train collisions per year (http://www.oli.org/statistics/collisions_casualties_year.htm); so corridors aren't foolproof. I can't think of an easier See & Avoid situation than a railroad crossing, but collisions are frequent.

Regarding 'detect and avoid', the current lack of this ability by most unmanned aircraft is one reason the FAA is 'reluctant' to give commercial unmanned aircraft more access to the NAS. Several groups are working on this issue including the ASTM F38 UAS Committee who is updating standards now for small unmanned aircraft. Last I heard the manned aircraft argument (I agree) was if the unmanned aircraft is sharing airspace it should be able to see and avoid at least as well as a pilot.

from a pure efficiency standpoint, I dont see a lot of gain in taking the pilots out of airline cockpits, though certainly it would save $$ for the airlines. But if small GA aircraft could be made virtually autonomous, even with some kind of remote monitoring, it might change the perception of GA among the masses. Of course this is talking about not what's possible today but perhaps forseeable in the next few decades...

The issue I have with small GA aircraft being virtually autonomous is the technology overlay and price! Look what is happening to the GA pilot population right now. Private pilot numbers are shrinking, there are significantly fewer new student pilots, and I think less than half become private pilots - who then find out how expensive it is to fly and stay current each year. In my area, to fly just 50 hours a year (rental) costs about $6,000 - a lot of money for most people. If we want to keep grassroot general aviation 'fun flying', we need access to affordable simple, non-complex, aircraft to fly. S-LSA was suppose to fill this market nitch and provide a surge of new sport pilots, but this has failed to happen due to who is actually buying most of the S-LSAs and the technology they want.

those are all good points, but in fact there are quite a few lower-cost options for pure fun flying -- ultralights, for example, or shared ownership of an older Cub or similar light aircraft -- but the costs go up when you want an aircraft to use as actual transportation. Making small aircraft virtually autonomous might attract more users, and more users would drive down costs. Plus staying current wouldn't be an issue.

S. Lanchester, it is interesting that you use the train analogy. Even with all of the car/train collisions that occur every year, there is no call to change the current situation. Everyone keeps harping that the automated airliner has to be "perfect", that even one accident would kill the move to automation. First of all, it is impossible to make a "perfect" airplane that never crashes. Even with pilots on board, this is not possible. Whereas the accident rate is very low for 121 carriers, it is not zero, and never can be, at least as long as there are humans involved.

Given all the growth that is forecast to happen, it will be hard for the mere human mind to keep up.

good point, David! also I wonder if the train example, though an interesting one, is comparable... those accidents, after all, take place at the road crossings, they are not happening between trains. If the train corridors were kept clear of crossings, there would be lots fewer issues.