Will it be politically acceptable to arm UAVs?
Another thing, would you want to be a infantryman supported by UAS doing CAS. With piloted aircraft you can (sometimes) talk directly you the pilot. You wouldn't be able to with a UAV.
The situation at the cutting edge may be incorrectly interpreted at control. A ground based controller will never have the same level of situational awareness as a pilot. If say a sensor broke (say an electrical joint failed due to vibration) you might failed to correctly ID a target. Another thread on PPRUNE talks about "friendly fire". These incidently would inevitably be much more common with UAVs.
As the the argument about a reduced need for communication, yes I agree, you could fly them with very little data transfer and they could fly in a "no comms" situation. But target ID and weapon release (particularly in the offensive role or with long range missiles) will not be entrusted to autonomous systems so a human bloke on the ground will have to OK weapon release. A shrewd enemy will seek to interupt communications.
Then there is the issue of infrastructure. A lot of extra infrastructure would be needed. Apart from the cost, these are now new targets for the enemy aircraft, missiles, UAVs, artillery, Special Forces or even terrorists.
Lastly.......On PPRUNE there is a thread relationg to the tragic Chinook crash in 1994. Many people, including myself, think that the most likely cause of this accident was faulty software. There is considerable evidence to support this. Software is almost impossible to test for 100% of all possible inputs, outputs, enviromental conditions, electrical noise in the system, interuptions to the power etc etc. Software based systems have failed spectacularly many times before. Consider...
Sea Wolf system software in Type 22 Frigates malfunctioning in the Falklands.
Computers "locking up" in the same conflict. This was a contributing factor in the loss of HMS Coventry.
Ariane 501. The software failed simply because a 64 bit number was put into a 16 bit register. Despite the hundred of millions of (whatever currency you like) put into the design and development, this still slipped through the net.
The loss of a pilotless Airbus. It flew right into the trees.
The loss of a US F22 due to a software malfunction.
Basically, allowing software to control things without having a human ON THE SPOT to make sure things are OK is just asking for trouble.
As for the issue of G limits, this is less of an issue now than it used to be. It a pilot losses conciousness for a second or two (and you would never experience more than 6G for longer this) it doesn't mean a crash. Modern instrumentation and controls (eg those being developed for the JSF) will tolerate transientry loss of pilot control.
Last edited by WE Branch Fanatic; 11th May 2002 at 14:44.