News story: First flight trials of Taranis aircraft
- Thread starter MoD_RSS
- Start date
Oh, yes.
In other news, I see Flight Global is reporting a proposed two-year delay to the F-35C's delivery. Before someone jumps all over me and says 'We're getting the -B', news like that makes me feel that pressing on with projects such as Taranis all the more important.
Lets give ourselves credit where credit is due - this is beyond a big step, we've jumped straight from theoretical powerpoint planes straight to the top of the class in a single leap and on a shoestring budget, a hell of an achievement considering the God knows how many $Billions the US is throwing at the same problems, and all 'our' own work - with a bit of help from the Australians.
To be sure, there is probably a long way to go yet, but we've started down the trail on a good footing. We're not at the 'Drone Wars' stage yet by any margin, but we are certainly a lot closer to the day when a flight of RAF attack drones from 617 (RPAS) Sqn head off to do their stuff thousands of miles away while the man in the loop sits back in his chair and feeds treats to the labrador of a certain hue.
If we do carry on with a joint stealth venture with the French- who to be fair have a very good aviation record - how do we go about resolving what they get to see and what we get to see etc ? Is it possible that due to our involvement with the US on F35 they might have a say in the Anglo-French collaboration in that they could be concerned that there might be leakage across projects ? Is getting involved with the US on stealth tech a good way to keep ourselves in the dark ?
That's the ultimate aim of the Taranis program though, autonomous operation.
And yes, it's an ethical issue, probably the greatest one in the military technology stakes since the development of nuclear weapons and one which is likely to attract a similar "anti" campaign, possibly an even greater "anti" campaign, which is why BAe don't advertise the autonomous element too widely.
Graculus
LE
They have a shite collaboration with good grace record and lots of previous for flouncing off stage left. I'd not be in the least bit shocked of they have a shafting planned.
SOI,
Equally, let's keep a sense of proportion. We've flown a single technology demonstrator from a land base.
In comparison the US have conducted autonomous dry AAR prods, carrier launches and traps, and have had the RQ-170 deployed on ops to multiple locations for almost a decade. And that's just what's in the open press!
Teranis is promising but I certainly don't see it as placing us 'straight to the top of the class in a single leap.'
No it isn't...and certainly not for the RAF. Don't mistake increased automation for autonomous.
Regards,
MM
We've had autonomous weapons for decades, it's just that they've been one way and we've called them cruise missiles or similar. At the simplest level - and hence first into service - loading up a mission plan for a UCAV to bomb a predefined DMPI is little different from loading up a plan for a Tomahawk to headbutt it.
Then we move onto recognising targets - and things start to get very difficult very fast. Well, radars and so on should be very doable, not many orphanages radiate like a SA-20 fire control radar after all. But as soon as you start to look for (say) tank shaped objects you start to throw more money down the black hole that has already consumed so much for so little reward. It is hard, hard, hard and even human beings - descendants of those good enough at pattern recognition to stay alive in environments full of camouflaged killing machines with lots of teeth - get it wrong far more often that you'd think. And human beings are also good at countermeasures, you're not looking for tanks, you're looking for tanks behind VISMODS and among decoys. If you're familiar with the way the Serbs took us to the cleaners in the FRY you should be looking rather sober right now.
I'm also very sceptical that true autonomy, the sort in the films, will ever be seen as worth the money by customers. If you're asking the PM to stake his political future on the DeathDrone 2000 not veering off course and bombing a school accidentally then the software effort will be ludicrously complex and expensive.
However, if you put a man in the loop (say as a backseater in a nearby fast jet) to press the commit button once he's had a look at some imagery with a big red cross on it then all of a sudden the problem becomes, in my opinion any road, quite doable. Have a flock of 5 or 10 UCAVs follow our notional bod in the jet and all of a sudden you can blow up a lot of shit quickly and still be reasonably sure nothing too unpleasant will occur.
I think you're quite right though in that the usual suspects will get their undergarments into a knot about it all.
Difficult to argue with one such as yourself on these issues MM, but the press releases from a few years ago definitely used the "A" word.
A prototype unmanned combat aircraft of the future, Taranis, has been unveiled by the MOD at BAE Systems’ facilities in Warton, Lancs. Named after the Celtic god of thunder, the concept demonstrator will test the possibility of developing the first ever autonomous stealthy Unmanned Combat Air Vehicle (UCAV) that would ultimately be capable of precisely striking targets at long range, even in another continent.
That's from a Quinetiq flyer.
Oots,
An excellent post which scratches the surface of the challenges of true automation as many consider it.
Even your SA-20 radar can be highly challenging to identify.
The laws of physics mean that an accurate fix can be very difficult to come from, particularly in 3-dimensions. Atmospherics, physical and urban geography, frequency harmonics and even the sun can all screw up the fix on an emitter. The same laws of physics mean that many threat radars possess almost identical parametrics to perfectly innocent emitters. EMCON measures such as blinking or 'shoot and scoot' tactics screw up your fixes further even before we get into modern Low Probability of Intercept techniques.
Then of course, there is the minor point that the SA-20 radar may be actually parked inside an orphanage!
Biscuits,
I genuinely never seek an argument and no offence was meant...honest!
There was - and to an extent continues to be - a great deal of loose language regarding 'autonomous ops' from contractors in particular. This was partly spin as they genuinely believed the military wanted such an attribute. They've now woken up to the fact that we don't as these simply play into the hands of our enemies. However, the press (and the anti-drone lobbyists) also seek out and exaggerate this concept.
Yet Western militarise are genuinely not seeking this capability as many perceive it. As an example, it's worth considering how the RAF paid to modify significant amounts of autonomous/'fire and forget' 'legacy' Brimstone to Dual Mode Seeker variants which introduced human input.
Autonomy has its place. Sometimes it can even be used in the application of kinetic effect. However, international law (rightly IMHO) severely limits the utility of such systems (for what we may consider civilised nations at least). Even allowing for the exponential increase in processing power, we remain a very long way away from a point where software lines can even approach the infinite degrees of judgement the Mk1 human brain is capable of.
The real value of combat RPAS will in my opinion be their ability to contribute to a mixed 'constellation' of networked manned and unmanned platforms, to share information, extend sensor and comm coverage, and a myriad of other ways.
Fully autonomous 'reusable cruise missile' strikes will at times probably be conducted, just as autonomous weapons can still be used today. However, just as such strikes have become increasingly rare and more restricted in their utility today, I see very few occasions in the foreseeable future where we'll launch genuinely autonomous weaponised systems.
Just like the 'D word', the 'A word' is inaccurate and should be discouraged!
Just my dribbling,
Regards,
MM
An excellent post which scratches the surface of the challenges of true automation as many consider it.
Even your SA-20 radar can be highly challenging to identify.
The laws of physics mean that an accurate fix can be very difficult to come from, particularly in 3-dimensions. Atmospherics, physical and urban geography, frequency harmonics and even the sun can all screw up the fix on an emitter. The same laws of physics mean that many threat radars possess almost identical parametrics to perfectly innocent emitters. EMCON measures such as blinking or 'shoot and scoot' tactics screw up your fixes further even before we get into modern Low Probability of Intercept techniques.
Then of course, there is the minor point that the SA-20 radar may be actually parked inside an orphanage!
Biscuits,
I genuinely never seek an argument and no offence was meant...honest!
There was - and to an extent continues to be - a great deal of loose language regarding 'autonomous ops' from contractors in particular. This was partly spin as they genuinely believed the military wanted such an attribute. They've now woken up to the fact that we don't as these simply play into the hands of our enemies. However, the press (and the anti-drone lobbyists) also seek out and exaggerate this concept.
Yet Western militarise are genuinely not seeking this capability as many perceive it. As an example, it's worth considering how the RAF paid to modify significant amounts of autonomous/'fire and forget' 'legacy' Brimstone to Dual Mode Seeker variants which introduced human input.
Autonomy has its place. Sometimes it can even be used in the application of kinetic effect. However, international law (rightly IMHO) severely limits the utility of such systems (for what we may consider civilised nations at least). Even allowing for the exponential increase in processing power, we remain a very long way away from a point where software lines can even approach the infinite degrees of judgement the Mk1 human brain is capable of.
The real value of combat RPAS will in my opinion be their ability to contribute to a mixed 'constellation' of networked manned and unmanned platforms, to share information, extend sensor and comm coverage, and a myriad of other ways.
Fully autonomous 'reusable cruise missile' strikes will at times probably be conducted, just as autonomous weapons can still be used today. However, just as such strikes have become increasingly rare and more restricted in their utility today, I see very few occasions in the foreseeable future where we'll launch genuinely autonomous weaponised systems.
Just like the 'D word', the 'A word' is inaccurate and should be discouraged!
Just my dribbling,
Regards,
MM
Similar threads
Latest Threads
-
-
-
-
-
How to prep for AOSB assault course
- Started by tertius
- Replies: 3