Navy F/A-18 Squadron Commander's take on A-I repeatedly...

12
1 archived as http://www.stealthskater.com/Documents/TopGun_04.doc (also …TopGun_04.pdf) => doc pdf URL-doc URL-pdf more Military topics are on the /Military.htm page at doc pdf URL note: because important websites are frequently "here today but gone tomorrow", the following was archived from https://www.thedrive.com/the-war-zone/35947/navy-f-a-18-squadron- commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers from the aforementioned website. Indeed, the reader should only read this back-up copy if it cannot be found at the original author's site. Navy F/A-18 Squadron Commander's take on A-I repeatedly beating a Human Pilot in Dogfight Everyone has an opinion when it comes to the stunning results of DARPA's AlphaDogfight trials. Now hear what the skipper of a fighter squadron thinks. by Commander Colin 'Farva' Price August 24, 2020 / The WarZone [Editor's note: To get up to speed, you can read all about the AlphaDogfight trials and their stunning outcome in this recent WarZone piece. You can also watch the final round where the A-I faced-off against the USAF F-16 Weapons Instructor and hear some commentary about the high-profile trials from those that worked on it for DARPA in the official video below.

Transcript of Navy F/A-18 Squadron Commander's take on A-I repeatedly...

Page 1: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

1

archived as http://www.stealthskater.com/Documents/TopGun_04.doc

(also …TopGun_04.pdf) => doc pdf URL-doc URL-pdf

more Military topics are on the /Military.htm page at doc pdf URL

note: because important websites are frequently "here today but gone tomorrow", the following was

archived from https://www.thedrive.com/the-war-zone/35947/navy-f-a-18-squadron-

commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is

NOT an attempt to divert readers from the aforementioned website. Indeed, the reader should

only read this back-up copy if it cannot be found at the original author's site.

Navy F/A-18 Squadron Commander's take on A-I

repeatedly beating a Human Pilot in Dogfight

Everyone has an opinion when it comes to the stunning results of DARPA's

AlphaDogfight trials. Now hear what the skipper of a fighter squadron thinks.

by Commander Colin 'Farva' Price

August 24, 2020 / The WarZone

[Editor's note: To get up to speed, you can read all about the AlphaDogfight trials and their stunning

outcome in this recent WarZone piece. You can also watch the final round where the A-I faced-off

against the USAF F-16 Weapons Instructor and hear some commentary about the high-profile trials

from those that worked on it for DARPA in the official video below.

Page 2: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

2

The recent 5-to-0 victory of an Artificial-Intelligence (A-I) pilot developed by Heron Systems over

an Air Force F-16 human pilot does not have me scrambling to send out applications for a new job.

However, I was impressed by the AlphaDogfight trials and recognize its value in determining where the

military can capitalize on A-I applications.

For most military aviators, it may be easy to scoff at the artificiality of the contest. I may have even

mumbled “Never would have happened to a Navy pilot…” Instead, I think it is important not to get

wrapped too much around the axle about the rules of the contest and instead focus on a couple of details

that really jumped out at me on the advantages that an A-I pilot would have over a human pilot.

For the contest setup, the argument about the death of the dogfight or that there is no need for within

visual range engagements anymore is a tired one. There was a pretty popular movie in the ‘80s about

that very argument. So I am not going to rehash it here. The fact is we still constantly train to dogfight

in the Navy or as it is more commonly referred to 'Basic Fighter Maneuvers' (BFM for short).

BFM is great airborne training for gaining an understanding of your energy state in relation to the

enemy and to exercise your situational awareness in a 3-dimensional space in a physically demanding

environment. An aviator has to understand how to aggressively maneuver their aircraft while at the

same time integrating their weapon systems to cue a weapon, assess the quality of the weapons track,

and determine if the trigger should be pulled to employ the weapon.

All at the same time, they must be preventing the enemy from accomplishing the same process. It is

a dynamic and stressful environment that creates better fighter pilots. I have yet to meet a pilot who is

an above-average BFM pilot but struggles in other mission sets.

There are multiple reasons why aircrew may find themselves at the merge with the enemy. But if

they do end up at the merge, the goal is always the same: take the first shot to kill the enemy before they

can shoot them.

This fact sometimes gets lost in training engagements. To maximize the training, the BFM fight will

often be taken to a “logical conclusion”. Even though each aircraft may trade shots early in the fight,

the two aircraft will keep fighting down to the hard deck till there is an obvious winner. Aircrew will

come to the debrief patting themselves on the back for the gun footage they have of the other aircraft.

But once the footage is played, they realize they absorbed the first shot well before their triumphant gun

pipper placement. The real-world logical conclusion could have been very different if they were

missing a wing or engine because of a missile impact.

The goal at the merge of achieving the first shot must be continually hammered home.

Still, the reality is that missiles do not always guide and fuze. Thus we extend fights to teach

aviators how to continue to survive or turn a defensive situation into an offensive one. The true sport of

fighter jet aviators is a guns-only BFM engagement.

A guns-only BFM engagement is a test of who can efficiently maximize their energy package and

capitalize on each merge. Much like chess, truly great BFM pilots are thinking 2-to-3 merges ahead and

not just reacting.

Page 5: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

5

It does not take much skill to put the aircraft’s lift-vector on the other aircraft and yank on the Gs. In

fact if in doubt, just doing that will take care of 75 percent of the fight.

But BFM is about being smoothly aggressive. Understanding the difference between when it is

necessary to max-perform the aircraft and when it is time to preserve or efficiently gain energy back is

key. In a tight turning fight, gaining a couple of angles at each merge can suddenly result in one aircraft

saddled in the other aircraft’s control zone working a comfortable rear quarter gun-tracking shot.

In true gamesmanship fashion, the guns-only BFM engagement was the setting for the

AlphaDogfight contest. So what jumped out at me about the engagements?

Three main points. First was the aggressive use of accurate forward quarter gun employment.

Second was the A-I’s efficient use of energy. Lastly was the A-I’s ability to maintain high-performance

turns.

Page 6: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

6

During BFM engagements, we use training rules to keep aircrew and aircraft safe. An example of

this is using a "hard deck" which is usually 5,000 feet above the ground. Aircraft can fight down to this

pretend ground level and if an aircraft goes below the hard deck, they are considered a “rocks kill” and

the fight is ended. The 5,000 feet of separation from the actual ground provides a safety margin during

training.

Another training rule is forward-quarter gunshots are prohibited. There is a high potential for a mid-

air collision if aircraft are pointing at each other trying to employ their guns. Due to the lack of ability

to train to forward-quarter gunshots, it is not in most aviators' combat habit patterns approaching the

merge to employ such a tactic. Even so, it would be a low probability shot.

A pilot must simultaneously and continuously solve for plane-of-motion, range, and lead for a

successful gun employment. It is difficult enough for a heart of the envelope rear-quarter tracking shot

while also concentrating on controlling a low amount of closure and staying above the hard deck . At the

high rates of closure normal for a neutral head-on merge, a gun envelope would be available for around

3 seconds. 3 seconds of intense concentration to track, assess, and shoot while at the same time

avoiding hitting the other aircraft. The Heron Systems A-I on several occasions was able to rapidly fine-

tune a tracking solution and employ its simulated gun in this fashion.

Page 7: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

7

Additionally, A-I would not waste any brain cells on self-preservation approaching the merge

avoiding the other aircraft. It would just happen. The tracking, assessing, and employing process for a

missile is not much different than the gun. I am pretty confident A-I could shoot a valid missile shot

faster than I can given the same data I am currently presented within the cockpit.

F-18 Super Hornet pilot in the cockpit about to launch off the deck.

The second advantage of A-I was its ability to maintain an efficient energy state and lift vector

placement. BFM flights certainly instill aviators with confidence in flying their aircraft aggressively in

all regimes of the flight envelope. However in today’s prevalent fly-by-wire aircraft, there is less

aircraft feel providing feedback to the pilot. It takes a consistent instrument scan to check the aircraft is

at the correct G, airspeed or angle-of-attack for the given situation.

Even proficient aviators have to use a percentage of their concentration (i.e., situation awareness) on

not over-performing or under-performing the aircraft. A-I could easily track this task and would most

likely never bleed airspeed or altitude excessively, preserving vital potential and kinetic energy while

also fine-tuning lift vector placement on the other aircraft to continue the fight if required.

Lastly is A-I’s freedom from human physiological limitations. During the last engagement, both

aircraft were in a prolonged two-circle fight at 9 Gs on the deck. A two-circle fight is also referred to as

a "rate fight'. The winner is the aircraft who can track its nose faster around the circle which is directly

proportional (disregarding other tools such as thrust vectoring) to the amount of Gs being pulled.

More Gs means a faster turn rate. 9 Gs is extremely taxing on the body which the pilot in the contest

did not have to deal with, either. A human pilot would have to squeeze every muscle in the legs and

abdominals in addition to focused breathing in order to not blackout. During training, I maintained 9 Gs

in the centrifuge for about 30 seconds. Then I went home and took a nap. And that was without being

shot at.

A-I does not care about positive or negative Gs. It will perform the aircraft at the level required.

Page 8: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

8

F/A-18F maneuvering hard

The truth is current aircraft have to be built to support the 'pile of human' sitting in it. The human

will always be the limiting factor in the performance of an aircraft.

I fight the jet differently now than I did as a junior officer when I was young and flexible. I have to

fight differently. I know what my capabilities are to get a consistent and repeatable shot with the little

bit of neck magic I have left to keep sight of the other aircraft.

The fact that in the contest, the A-I had perfect information at all times, and rules of engagement

were not a factor, are not inconsequential details. I recognize that providing the amount of data and

sensor fusion the A-I would require to perform at the same level in a real aerial engagement (one that

does not take place in cyberspace) is not a small undertaking and still a bit in the future. The Rules of

Engagement discussion could fill up the syllabus for the entire semester of an ethics class and will

always be a touchy subject with regards to A-I's involvement in war.

I am not an engineer nor an ethics professor. Yet as a pilot, I am intrigued. A computer model was

able to react to the movements of a human pilot and effectively employ weapons. During the 5

engagements, the A-I had 15 valid gun employments and the human pilot had zero. These results also

hint at the A-I’s ability to avoid being shot while effectively employing its own weapons.

Page 9: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

9

Farve and his fellow Black Knights of VFA-154 flying and fighting in the age of COVID.

An A/I-enhanced weapon’s employment system in my aircraft? I am not ready for the Terminator's

Skynet to become self-aware. But I am certainly ready to invite A-I into the cockpit. Hell, I am only a

voting member as far as the flight controls are concerned in the Super Hornet anyways. If I put a control

input in that is not aerodynamically sound (i.e., could result in a departure from controlled flight), the

flight control system will not move the control surface or will move a different surface to give me the

movement I am requesting. Who is flying Who?

So if tomorrow my 7-year-old daughter decides she wants to become a Naval Aviator, I am not

going to shoot down the notion and go on a rant about the last generation of fighter pilots. I know there

will be a Navy jet for her to fly.

My future grandchildren, however? Saddle up kids and prepare yourself for some of Grandad’s wild

tails of the greatest flight in Naval Aviation: the one-hour BFM cycle back to the Case One shit-hot

break.

Those were the days!

Contact the editor: [email protected]

Page 10: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

10

Reader Comments

1. SeaDemon

The amazing thing about an A-I is that its intelligence and ability do not die when its vehicle is lost.

If the A-I is receiving data from the fights that it loses, it can use that information to recognize a similar

situation, attempt another solution, and then add that experience to its decision tree. That and an A-I's

understanding of itself as a network of nodes will make it an incredibly tough opponent as this tech

matures.

2. falcon642

"I am pretty confident AI could shoot a valid missile shot faster than I can, given the same data I

am currently presented within the cockpit.

BVR missile engagement and prosecution of timelines is an area where A-I has some serious

potential. The A-I can constantly calculate the odds of its own missile hitting and of the enemy missile

hitting it. So if the A-I can calculate that if it waits another 0.7 seconds before firing its missile and

breaking, then its chance of getting a kill goes up by 5% while its chance of getting killed only goes up

1%, then the A-I is probably going to take that increased risk to get the increased reward.

Fighter Pilot Podcast did an awesome episode on air-to-air timelines. Given all the things that go

into developing a timeline, it's hard not see how A-I would really excel at BVR missile engagements as

well.

Employing an air-to-air missile beyond visual range (BVR) is not simply a matter of detecting the

target and pulling the trigger. A host of factors demand consideration. What is the threat and what

kinds of missiles does it employ? From what speed and altitude? With this in mind, we can plan to

employ so that our missile hits the threat before the threat’s missile hits us.

https://www.fighterpilotpodcast.com/episodes/071-air-to-air-mission-planning/

3. stanleywinthrop

At its heart, BFM is about energy management. (Nose position used to be an important factor. But

HOBS weapons have diminished that factor.). When you boil it down energy management is a dynamic

math problem with 2 factors: kinetic and potential energy. An airplane that does a better job of

managing both those factors in relation to an opponent will always have an advantage. It's no surprise

that A-I has already damn near mastered that side of BFM. The other parts will no doubt come with

time.

4. TurboDan813

C.W. "Mover" Lemoine (former Super Hornet, Viper, and current T-38 pilot) did a great review

linked below. "Ate" Chuet (former Rafale-M and Super E driver with the French Navy) did an excellent

video too. But you need to toggle the closed captioning from French to English.

Ate's Take: https://www.youtube.com/watch?v=v4PNGkif1nw

Mover's Take: https://www.youtube.com/watch?v=ziCQqmEllZo

Page 11: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

11

5. JS Dreyer

I think the next phase will be AI-assisted tactical jets that also fly with A/I-controlled drones (i.e.,

loyal wingman). The pilot (or more likely his backseater) will vector the drones out ahead to manage

threats and the human-piloted aircraft will only engage as a last resort.

6. Malakas Keratas

Is there any development on something like A-I augmented assistance for pilots for things like

shooting and maneuvering as a stop gap? Between now and the pilotless A-I future?

The pilot would still have overriding control including pulling the trigger and designating the target

as well as overriding any piloting assistance. The pilot would still essentially guide the cross hairs onto

the target. The A-I would just seamlessly make many micro adjustments to optimize the pilots

intentions.

A/I-augmented piloting combined with 3D thrust vectoring would allow pilots to pull off more

aggressive maneuvers with higher G limits that are not physically possible atm. The higher g-forces

could be sustained by something like auto reclining seats that self-adjust and are capable of lying

completely flat as the A-I assists maneuvering during tighter turns. With helmets like the new F-35

HMDS, the pilot could still maintain situational awareness as he is momentarily reclined during an

extreme maneuver.

7. leovinus2001

First of all, thank you for your level-headed assessment and your service!

"I am not an engineer nor an ethics professor. Yet as a pilot ... ..."

That's fine. Personally, I come in from the other side as a scientist and engineer who has been

building, analyzing, deploying, and cursing various machine learning (ML) models during the last 30+

years. And still on a day-to-day basis.

While the "A-I" is mostly marketing which should have died in the 90's when Japan failed at the

Prolog deterministic approaches, what is called A-I these days is mostly back-propagation trained

networks based on mathematics from the 80s (Rumelhart, Hinton) but pumped up like you would not

believe. Brute force plus some newer algorithms (auto-differentiate, LSTM) saved the day and opened

up venues.

Having said that, what I have hinted at various times, asked for, and would hope that JAIC et al

would realize is that you get the most effective solutions when you have pilots like yourself talk to

AI/ML experts like myself and build some kick-ass prototypes.

DARPA and the ML community have been discussion since the 70s. A nice example is the ICASSP

conference which always had it "underwater acoustics" session next to the speech recognition session

where often an USAF man would be punting F-16 cockpit noise CDs to foster noise robust research.

When we cooperate like that, the combo will be great.

Page 12: Navy F/A-18 Squadron Commander's take on A-I repeatedly ...commanders-take-on-ai-repeatedly-beating-real-pilot-in-dogfight on August 25,2020. This is NOT an attempt to divert readers

12

For me, one takeaway from the recent test is that there are training sets of data based on F-16 flying

data which have gotten us to the point that we can beat a human expert. Nice! That is comparable to the

first chess programs beating the world champion (a milestone). It does not matter whether the algorithm

is reinforcement learning or seq2seq. If you have a track of positions/attitudes/velocities in 3D to work

with, you can do amazing things.

"Yet, as a pilot, I am intrigued."

Good to hear. As ML expert, so am I. Where do we go from here and how can we help? We can

make the pilots more effective. We can save lives. We need to drink beers together to find ideas to help

you. Put me in the backseat of your Hornet and I guarantee I'll come up with one novel insight or I pay

for the ride.

"Still, the reality is that missiles do not always guide and fuze ..."

Well, complain to the manufacturer of the missiles and let them hire better engineers who build

better classifiers which track the 'dot' in front of you more effectively. While I can run the number

crunching at home, the data to help you guys is not out there.

if on the Internet, Press <BACK> on your browser to return to

the previous page (or go to www.stealthskater.com)

else if accessing these files from the CD in a MS-Word session, simply <CLOSE> this

file's window-session; the previous window-session should still remain 'active'