USAF/USN 6th Gen Fighters - F/A-XX, F-X, NGAD, PCA, ASFS News & Analysis [2008- 2025]

Status
Not open for further replies.
It is the JSF program all over again. Who knows who will win this time round, especially with NG concentrating on the Navy F/A-XX.
 
Where did you hear/see that information dark sidius? I am highly interested. The SR-72 has gone very quiet as of late, I was worried that Lockheed had cancled the program.
 
red admiral said:
If everything is really automated, then why do you need an extra crewman dedicated to this?
Click to expand...

Because getting to 100% automation is massively more difficult than to 80% automation and having a WSO work with it.

A 2-man crew is the difference between an 80% solution for CCA and shooting for 100% or bust. As Scott Kenny put it, nobody has demonstrated an autonomous CCA or anything remotely approaching it. A bit of humility about our technological capabilities would go a long way in getting some form of CCA capability now rather than leaving it in development hell.
 
Scott Kenny said:
But it hasn't been, at least not in any public discussions.

Testing of MUMT with Apache Es and MQ1Cs has shown that the current generation of drones require far too much pilot/operator input for someone to fly an aircraft and fly the drone at the same time.

So unless the current CCA designs are at a point where they require as much operator interaction as a JDAM, you're going to need a back seater to quarterback the drones while a pilot keeps the quarterback safe.
Click to expand...

Btw, is there any concept of operations in public domain that shows envisaged operational scenarios in detail?

In particular for ground launched / reusable "loyal wingman" in the likes of MQ-28. I have a hard time to imagine how multiple vehicles gonna work in concert with a high performance fighter, from take-off to landing.
 
DrRansom said:
As Scott Kenny put it, nobody has demonstrated an autonomous CCA or anything remotely approaching it. A bit of humility about our technological capabilities would go a long way in getting some form of CCA capability now rather than leaving it in development hell.
Click to expand...
There's been plenty of live demonstrations of autonomous aircraft taking task based commands

More the thing is that "autonomous" isn't some binary thing; it very much depends on what functionality the system has, and then which bits do you want to automate and how much; and what the risk appetite is for someone to sign off the safety case, with what constraints
 
Has anyone demonstrated anything like CCA activity in a modern battlefield? It takes some effort to make an autonomous demonstration, but demonstrations really mean nothing. It takes far far more effort to make robust autonomy. Autonomous cars were demonstrated a decade ago, we're nowhere near autonomous cars. Compared to the CCA, autonomous cars have had far more money and better engineer talent working on them, for very little practical result.

Hence the WSO: a WSO lowers the development target threshold compared to single-piloted aircraft.
 
VTOLicious said:
Btw, is there any concept of operations in public domain that shows envisaged operational scenarios in detail?
Click to expand...
I haven't seen one, but most of the folks here are better plugged in than I am to hear about or find an open source CONOPS.

I'm pretty sure the eventual goal is for one manned plane leading an entire strike package of ~12x CCAs, but as far as I know the software isn't there yet. At least not reliably there.


DrRansom said:
Has anyone demonstrated anything like CCA activity in a modern battlefield? It takes some effort to make an autonomous demonstration, but demonstrations really mean nothing. It takes far far more effort to make robust autonomy. Autonomous cars were demonstrated a decade ago, we're nowhere near autonomous cars. Compared to the CCA, autonomous cars have had far more money and better engineer talent working on them, for very little practical result.
Click to expand...
I know the Army has been doing MUMT with Apache Es and Gray Eagles, I don't know if that has been in combat or what.


DrRansom said:
Hence the WSO: a WSO lowers the development target threshold compared to single-piloted aircraft.
Click to expand...
The Army MUMT has shown that at least Predator level drones need too much operator input for one person to both fly a plane and fly the drone.
 
VTOLicious said:
Btw, is there any concept of operations in public domain that shows envisaged operational scenarios in detail?

In particular for ground launched / reusable "loyal wingman" in the likes of MQ-28. I have a hard time to imagine how multiple vehicles gonna work in concert with a high performance fighter, from take-off to landing.
Click to expand...
They won't all work in concert from take-off to landing. Some will meet at way points. Munitions can also be launched from over the horizon from one node, targeting directed by a second node, all managed by piloted aircraft. Line of sight from the piloted aircraft to the targeting node or munition is all that's required.
 
NeilChapman said:
They won't all work in concert from take-off to landing. Some will meet at way points. Munitions can also be launched from over the horizon from one node, targeting directed by a second node, all managed by piloted aircraft. Line of sight from the piloted aircraft to the targeting node or munition is all that's required.
Click to expand...
That's harder to do over the Pacific. Only so many places to launch from, so it's quite possible that the CCAs are going to fly with the manned plane from launch to recovery.
 
DrRansom said:
Hence the WSO: a WSO lowers the development target threshold compared to single-piloted aircraft.
Click to expand...
If you want a human to help the automation, then why do they need to be in an tacitcal aeroplane vs having many people sitting comfortably in a big airliner or on the ground?
 
Scott Kenny said:
Simpler communication paths and shorter comms lag.
Click to expand...
Why are those an issue for something sat 200-400nm back when it's not for Reaper Ops the other side of the world? And this is for something more automated than a remotely piloted system like Reaper

Why is having one WSO enough? Why not have a crew of 3, or 4, or more?
 
red admiral said:
Why are those an issue for something sat 200-400nm back when it's not for Reaper Ops the other side of the world? And this is for something more automated than a remotely piloted system like Reaper
Click to expand...
Because Reapers are not flying in/into denied airspace like the NGAD and CCAs will be, so you don't care about a 3sec comms delay.




red admiral said:
Why is having one WSO enough? Why not have a crew of 3, or 4, or more?
Click to expand...
Because having a crew of 3+ means a much bigger aircraft, barring doing something weird like the EA-6B Prowler (5000lbs heavier than the A-6E). The E-11A BACN comms router has a crew of at least 4 and is a 100klb MTOW aircraft. Doing that in an NGAD compliant airframe is likely to make a 120klb aircraft.
 
I would think that in a future where a major war could potentially include anti-satellite warfare and widespread EW and jamming that you wouldn't want to rely on any kind of remote WSO in a combat situation.
If a Reaper gets jammed its not the end of the world, a manned platform can take over. But a strike-fighter is totally different. And two pairs of eyes work better in the sky.
 
That is true Hood, there is no better way of scanning the sky than with the MK.1 Eyeball especially if the radar is getting jammed by hostile electronic warfare jammers.
 
Perhaps Lockheed had realised that the SR-72 was too advanced for the current time technology wise and has gone back to the drawing boards.
 
Scott Kenny said:
That's harder to do over the Pacific. Only so many places to launch from, so it's quite possible that the CCAs are going to fly with the manned plane from launch to recovery.
Click to expand...

What is your definition of "with"? Line of site at 60'k is ~300 miles.
 
red admiral said:
If you want a human to help the automation, then why do they need to be in an tacitcal aeroplane vs having many people sitting comfortably in a big airliner or on the ground?
Click to expand...

You're seriously asking that question? This is the 21st century we're talking about - all communication links (satellite, airborne nodes) are subject to violent interdiction and anything big and non-stealthy can't get within 200nm of the front, lest it get sniped by some hostile stealth fighter with a long-range AAM and super-cruise.

As for 2 pilots versus 3 or 4, that's just being difficult. As far as I see it, 2 >> 1, for the reasons mentioned above and based on the Army experience as related by Scott Kenny. Going to 3 or 4 would be overkill - unless autonomy development completely fails, at which point the whole discussion is moot anyway.
 
From last year. Pretty ballsy from Kendall to confidently make such claim when 3 days before it was said NGAD must be fielded at least "a month before our competitors". Two contradictory statements.
breakingdefense.com

China 'on track' for 6th-gen fighter, US Air Force needs to get there first: ACC chief - Breaking Defense

America's Chinese "frenemies" are "not having a debate over the relevance of six-gen air dominance. And I can also tell you they're on track," said Air Combat Command head Gen. Mark Kelly.
breakingdefense.com breakingdefense.com
defensescoop.com

Kendall: Air Force’s next-gen fighter will be ‘better than anything' China can produce

The Air Force plans to team its Next-Generation Air Dominance platform with robotic wingmen.
defensescoop.com defensescoop.com
 
Scott Kenny said:
"stuck launching from the same bases as manned aircraft because there's no US-controlled or US-friendly airfield closer"
Click to expand...
That doesn't require they take off at the same time.They can also take off from a multitude of locations and meet at a suitable waypoint enroute. Neither requires satellite communications to achieve.
 
NeilChapman said:
That doesn't require they take off at the same time.They can also take off from a multitude of locations and meet at a suitable waypoint enroute. Neither requires satellite communications to achieve.
Click to expand...
Yes, the CCAs may be launching from a physically separate base from the manned aircraft. Depending on where and how the control handoff occurs, they may be able to meet up en route.

I will still bet that it will require functional satcom links from CCA takeoff to CCA handoff, just like flying a Reaper needs satcom.
 
DrRansom said:
You're seriously asking that question? This is the 21st century we're talking about - all communication links (satellite, airborne nodes) are subject to violent interdiction
Click to expand...
It's a good job that our crewed aircraft aren't also critically dependent on those same communication links. Oh wait...
 
Scott Kenny said:
But it hasn't been, at least not in any public discussions.

Testing of MUMT with Apache Es and MQ1Cs has shown that the current generation of drones require far too much pilot/operator input for someone to fly an aircraft and fly the drone at the same time.

So unless the current CCA designs are at a point where they require as much operator interaction as a JDAM, you're going to need a back seater to quarterback the drones while a pilot keeps the quarterback safe.
Click to expand...

Grey Eagle isn't a CCA, it's a remotely piloted aircraft. It is at least two generations behind the state of the art, and requires near constant management in the loop control to do anything. The mission of the type also did it no favors, requiring constant positive control for counterinsurgency ISR and strike missions. This is akin to saying we couldn't possible make CEC work because Link 11 didn't have the throughput or data quality to allow it—the comparison is decades outdated.

The CCAs the Air Force and Navy are working on developing are intended to be vastly more autonomous, so the aircrew (which will probably be one individual) doesn't have to direct the aircraft to perform much of its mission. Instead they will act as extensions of the aircraft, able to employ sensors and weapons tens of kilometers from the piloted aircraft. The majority of direct control pilots will have over CCAs will likely be effectively modal—i.e. formations or EMCON modes. But for weapons employment they should be able to simply make the decision to shoot and their aircraft in concert with the CCAs will make the decision which one actually takes the shot.
 
red admiral said:
It's a good job that our crewed aircraft aren't also critically dependent on those same communication links. Oh wait...
Click to expand...
But a manned aircraft has a brain - or two brains in a two-seater - to complete the mission under its own initiative if needs be. A UAV without a datalink is just a flying airprox hazard until the fuel runs out. Maybe one day AI will enable a UAV to carry out its mission autonomously without any datalink, but in the here and now that's not possible.
Of course GPS-equipped weapons are just as vulnerable too - perhaps too many eggs in one basket.
 
Scott Kenny said:
But it hasn't been, at least not in any public discussions.

Testing of MUMT with Apache Es and MQ1Cs has shown that the current generation of drones require far too much pilot/operator input for someone to fly an aircraft and fly the drone at the same time.

So unless the current CCA designs are at a point where they require as much operator interaction as a JDAM, you're going to need a back seater to quarterback the drones while a pilot keeps the quarterback safe.
Click to expand...

I think the goal of the Skyborg program is to get CCAs to the point where the pilot only assigns general behaviors and weapons release authority. Individual aircraft might be designated as offensive or defensive, active emitters or passive, cautious or aggressive/expendable, weapons free/tight/pilot authorized only, etc. Then the CCAs assume a formation around the manned aircraft and reorient with it as it changes speed and direction or as behavior settings are altered. That is how I envision it working. If more attention is needed, then it does seem a dedicated operator is needed, but I think the goal is a higher level of pre programmed behaviors more akin to a living wing man and/or second flight element.
 
Scott Kenny said:
That's harder to do over the Pacific. Only so many places to launch from, so it's quite possible that the CCAs are going to fly with the manned plane from launch to recovery.
Click to expand...

Other options might be runway independent CCAs or a manned NGAD platform that operates from further out meet up with CCAs of a shorter range launched from closer airbases. It seems likely the manned fighter is going to be more constrained by runway length/size/weight limit than the CCAs.
 
totallyaverage said:
Grey Eagle isn't a CCA, it's a remotely piloted aircraft. It is at least two generations behind the state of the art, and requires near constant management in the loop control to do anything. The mission of the type also did it no favors, requiring constant positive control for counterinsurgency ISR and strike missions. This is akin to saying we couldn't possible make CEC work because Link 11 didn't have the throughput or data quality to allow it—the comparison is decades outdated.

The CCAs the Air Force and Navy are working on developing are intended to be vastly more autonomous, so the aircrew (which will probably be one individual) doesn't have to direct the aircraft to perform much of its mission. Instead they will act as extensions of the aircraft, able to employ sensors and weapons tens of kilometers from the piloted aircraft. The majority of direct control pilots will have over CCAs will likely be effectively modal—i.e. formations or EMCON modes. But for weapons employment they should be able to simply make the decision to shoot and their aircraft in concert with the CCAs will make the decision which one actually takes the shot.
Click to expand...
Yes, that's the end goal they're working towards.

Parts of the AI is there, 15 years ago there was a demonstration where a target was assigned to a flight of 4 drones and the drones sorted the attack out on their own, assigning planes to decoy, SEAD, and actual target autonomously.

Not all of the AI is there yet.


Josh_TN said:
I think the goal of the Skyborg program is to get CCAs to the point where the pilot only assigns general behaviors and weapons release authority. Individual aircraft might be designated as offensive or defensive, active emitters or passive, cautious or aggressive/expendable, weapons free/tight/pilot authorized only, etc. Then the CCAs assume a formation around the manned aircraft and reorient with it as it changes speed and direction or as behavior settings are altered. That is how I envision it working. If more attention is needed, then it does seem a dedicated operator is needed, but I think the goal is a higher level of pre programmed behaviors more akin to a living wing man and/or second flight element.
Click to expand...
That's certainly where I think it needs to be, so you can have one manned plane as the quarterback for the entire strike package of EW and recon and strike and fighter escort and SEAD/DEAD.

But parts of that level of autonomy just aren't there yet. I believe the programming term is "it has turned out to be a wicked problem."


Josh_TN said:
Other options might be runway independent CCAs or a manned NGAD platform that operates from further out meet up with CCAs of a shorter range launched from closer airbases. It seems likely the manned fighter is going to be more constrained by runway length/size/weight limit than the CCAs.
Click to expand...
You need to look at what runways are available in the Pacific, and how many islands big enough for a runway are within the A2AD zone.
 
Scott Kenny said:
You need to look at what runways are available in the Pacific, and how many islands big enough for a runway are within the A2AD zone.
Click to expand...

Every WesPac airbase is inside ballistic missile range. But the lower the runway requirements of CCA, the larger number of targets that have to be hit and the more hits to runways surfaces are necessary to make them inoperable. It strikes me that CCA will have less stringent runway requirements, where as I suspect NGAD manned is going to be large enough it can only stage from major airbases, possibly out of theater.
 
Hood said:
Of course GPS-equipped weapons are just as vulnerable too - perhaps too many eggs in one basket.
Click to expand...

How so? GPS equipped weapons use GPS to update an inertial navigation system. If GPS is jammed, spoofed, etc the effect on the weapon is negligible.
 
quellish said:
How so? GPS equipped weapons use GPS to update an inertial navigation system. If GPS is jammed, spoofed, etc the effect on the weapon is negligible.
Click to expand...

Negligible is perhaps a little strongly worded. Degraded accuracy in an INS/PGS only guided weapon would be really limiting against harden targets. On the other hand there are numerous weapons that use terminal seekers on top of GPS which likely would be marginally affected-something like JASSM or SDB II is likely getting close enough to its target to be within its seekers FoV.
 
Status
Not open for further replies.

Similar threads

Back
Top Bottom