The pilots were only on board as a precaution;
www.lep.co.uk/news/regional/first-unmanned-flight-completed-1-5669134
I can't wait for this technology to get into motor cars.
|
Already been done, all you need is a military drone, remove the missiles and bombs and fit a few seats. :-)
Driverless cars are legal in California.
Last edited by: Old Navy on Wed 15 May 13 at 10:31
|
In a normal commercial aeroplane, there is a failsafe in the event of the pilot being taken ill/dying suddenly. The co-pilot.
What would be the failsafe in the event of a complete system failure in a pilotless aircraft driven remotely? If it's to have a pilot on board, then what's the point?
|
|
Flying without aircrew isn't anything new. It's been done for years in the military world. What will take a longer time is the clearing all the objections for passenger aircraft. I suspect we will be all long dead and buried before it happens.
|
>> passenger aircraft.
That's what I was talking about.
|
|
I know you were. I was just adding my thoughts.
|
|
I can remember witnessing missiles being fired at pilotless full sized aircraft in the 1960s.
|
Rather oddly, it's easier to make a fully autonomous aircraft than a fully autonomous car, despite the aircraft being obliged to navigate in three dimensions rather than two.
Things don't walk / ride / drive out in front of moving aircraft.
The only thing an aircraft is likely to run into up there is another aircraft[1] and these can be easily detected at sufficient range to be avoided without too much fuss.
Teams of blokes in yellow jackets do not turn up to dig up the sky, cone sections of it off or make it one way only. If an aircraft is diverted it will be to another airport rather than off round a set of back streets crammed with stationary traffic.
Aircraft are only obliged to deal with fixed obstacles during takeoff and landing, unless there's been a right pig's ear somewhere. This always occurs at points which are liberally equipped with things designed to aid automation of the process.
Accuracy of positioning is only necessary to "somewhere around the right place" in flight and within a few feet while taking off and landing, rather the the inches perfect positioning required of a car.
Etc....
In fact the sky is a very big place and if aircraft were allowed to find their own way around it rather than being assigned to fixed bits of it for the convenience of human controllers, they'd be far less likely to ever see another one outside the immediate vicinity of an airport.
Most of the bits are already in place. It's already perfectly possible for a commercial pilot's sole intervention in a flight to be: press takeoff button, engage autopilot, press land button. Tying those bits together with a system that can also take avoiding action based on the radar information and/or TCAS negotiation and handle flightplan updates from the ground is the missing piece of the puzzle.
The difficult bit is the "what happens when it goes wrong?" part. Here you have to get a bit ruthless. For every major mechanical failure where the actions of the crew saved the plane, there are probably four or five where there was nothing they could do and the aircraft went in.
More importantly, for every "the pilot saved it" event, there are several where the crew's screwing up of the routine work resulted in the loss of an aircraft.
As importantly, more than a few of those saves by a human pilot were saving the result of an earlier human pig's ear.
Thus, having fully autonomous aircraft would almost certainly result in fewer large aluminium tubes ploughing into the turf than we have now.
[1] Ok, I'll give you that one, but Austrian skydivers moving at supersonic speeds are actually quite rare.
|
|
Nothing new-I remember years ago a video of a BA 747 from Heathrow to India where the computer handled the taxi out to the runway,take-off,flight,landing and taxi in to the terminal.
|
|
I can remember from the early days of the Glasgow / London shuttle flights, landing at Heathrow in poor visibility and the pilot apologising for the (slight) bump on touchdown as it had been a fully automatic landing.
|
>> Most of the bits are already in place.
>> It's already perfectly possible for a commercial pilot's sole intervention in a
>> flight to be: press takeoff button, engage autopilot, press land button.
No, it's not. No current passenger aeroplane can perform an automatic take-off.
Having intervened against the Airbus autopilot and autothrust more times than I can count, I'll be the last person to get on a pilot-less aircraft.
|
>> Rather oddly, it's easier to make a fully autonomous aircraft than a fully autonomous car,
>> despite the aircraft being obliged to navigate in three dimensions rather than two.
Rather oddly that is completely wrong. Its far easier to make fully automated cars. So easy they currently exist, in practical todays terms its a complete doddle. You have much closer reference to solid objects (kerbs, walls, street furniture,) to enable short range ultrasonic and infra red detection and guidance. Streets and street furniture can be additional equipped to guid and control, speeds are much slower, and you only have to control in two dimensions.
As for people and cars dancing out in front of you, they only do it at slow speeds (in electronic control terms) and electronically controlled cars will be better at avoiding them. You also don't have to land a car at 150 mph plus in a 50 mph crosswind, and torrential rain.
|
|
Why is it called autonomous? It doesn't have an independent ability to do what it likes, I thought the point is the aircraft is told where to go?
|
ASTRAEA - autonomous flight - Old Navy
