THE ACCURATERELOADING.COM AVIATION FORUM

Go
New
Find
Notify
Tools
Reply
  
AI Beats Top Air Force Pilot
 Login/Join
 
one of us
posted
Had to happen sooner or later. One step closer to removing the man from the cockpit.

quote:

AI Slays Top F-16 Pilot In DARPA Dogfight Simulation

"It's a giant leap," said DARPA's Justin (call sign "Glock") Mock.

By THERESA HITCHENS

on August 20, 2020 at 6:03 PM

Heron Systems’ AI pilot scores the final kill against AF F-16 instructor “Banger” in AlphaDogfight Trials

WASHINGTON: In a 5 to 0 sweep, an AI ‘pilot’ developed by Heron Systems beat one of the Air Force’s top F-16 fighter pilots in DARPA’s simulated aerial dogfight contest today.

“It’s a giant leap,” said DARPA’s Justin (call sign “Glock”) Mock, who served as a commentator on the trials.

AI still has a long way to go before the Air Force pilots would be ready to hand over the stick to an artificial intelligence during combat, DARPA officials said during today’s live broadcast of the AlphaDogfight trials. But the three-day trials show that AI systems can credibly maneuver an aircraft in a simple, one-on-one combat scenario and shoot its forward guns in a classic, WWII-style dogfight. On the other hand, they said, it was an impressive showing by an AI agent after only a year of development. (As I reported earlier this week, the program began back in September last year with eight teams developing their respective AIs.)

Heron, a small, female- and minority-owned company with offices in Maryland and Virginia, builds artificial intelligence agents, and is also a player in DARPA’s Gamebreaker effort to explore tactics for disrupting enemy strategies using real-world games as platforms. The company beat eight other teams, including one led by defense giant Lockheed Martin — which came in second in the AlphaDogfight “semi-finals” that pitted the AI pilots against each other this morning.

Heron’s team did a live-stream Q&A on Youtube. “Even a week before Trial 1, we had agents that were not very good at flying at all. We really turned it around, and since then we’ve been really number one,” said Ben Bell, Heron’s co-lead for the project. The team intends to publish later this year some of the details about its reinforcement learning process for the AI, he said.

The trials were designed as a risk-reduction effort for DARPA’s Air Combat Evolution (ACE) program to flesh out how human and machine pilots share operational control of a fighter jet to maximize its chances of mission success. The overarching ACE concept is aimed at allowing the pilot to shift “from single platform operator to mission commander” in charge not just of flying their own aircraft but managing teams of drones slaved to their fighter jet. “ACE aims to deliver a capability that enables a pilot to attend to a broader, more global air command mission while their aircraft and teamed unmanned systems are engaged in individual tactics,” the ACE program website explains.

As tensions rise, ISR demand will increase exponentially as will the need for kinetic response to threats.

From GEN. FRANK GORENC, USAF (RET.)
Heron Systems’ AI was extremely aggressive in the games, with its AI pilot consistently able to turn and score killing hits on the simulated F-16 piloted by an unnamed Air Force pilot, with the call sign “Banger,” a graduate of the Air Force’s highly selective Weapons School at Nellis AFB. The AI exhibited “superhuman aiming ability” during the simulation, Mock said.

While the trials were not in anyway “definitive” of an AI pilot’s future capabilities or even its viability, Mock said, at the same time “what we saw was that in this limited area, in this specific scenario, we’ve got AI that works.”

DARPA intends to take the simulator used in the trials, and the simulations, to Nellis, where other Air Force pilots can take a stab at trying to beat AI pilots. Next steps will be to move on to testing AI pilots’ capabilities to perform other types of aerial combat missions.

Somewhere, the infamous Red Baron is no doubt laughing in amazement.


___________________

Just Remember, We ALL Told You So.
 
Posts: 22442 | Location: Occupying Little Minds Rent Free | Registered: 04 October 2012Reply With Quote
one of us
posted Hide Post
Yes it was bound to happen. I hope it is used to strengthen pilot capabilities and provide an increased measure of safety. Pilots in today's fighters can get seriously over-tasked. Having a capable and reliable AI co-pilot can relieve the mission commanders of distractions and help them concentrate on mission strategy. I'm excited about the drone management possibilities too. Just think of a squadron of fighters led by one aircraft with a human pilot. The drone fighters could be re-configured older F-16's with the cockpit and pilot protective armor removed. Maybe strengthened to handle 12 + G's. They could outmaneuver anything flying of human hands. The flying "tic-tacs" are another matter! Wink
 
Posts: 2834 | Location: SC,USA | Registered: 07 March 2002Reply With Quote
one of us
posted Hide Post
With Boeing and Krakos testing Loyal Wingman UCAVs, the combat drone isn't too far away now that aerial combat AI has matured.


___________________

Just Remember, We ALL Told You So.
 
Posts: 22442 | Location: Occupying Little Minds Rent Free | Registered: 04 October 2012Reply With Quote
one of us
Picture of Todd Williams
posted Hide Post
I'm not completely sure how to feel about this.

On the one hand, and obviously the most important aspect, anything that increases our combat effectiveness is a step in the right direction, reducing risk to our personnel and equipment.

On the other hand, as a former fighter pilot myself, I can't help but feel diminishment in the loss of the man to man contest of Ariel combat, often almost considered gentlemanly, although certainly a high stakes game with consequences of death or captivity very possible. That's the way Richthofen viewed it, and I think most of the guys I flew with held at least some semblance of respect for our opponents.
 
Posts: 8107 | Registered: 09 January 2011Reply With Quote
one of us
Picture of TCLouis
posted Hide Post
And just think how much money will be saved not having to support a human on board.
Weight savings can go toward additional ordnance.

A10 upgraded to support troops on the ground that actually hold the contested real estate.



Don't limit your challenges . . .
Challenge your limits


 
Posts: 4029 | Location: TN USA | Registered: 17 March 2002Reply With Quote
one of us
posted Hide Post
If you consider how much engineering, air frame design, weight and cost that goes into keeping a man in the cockpit, moving towards a mostly UCAV fighting force makes a lot of sense on many, many levels.

With all of the latest in sensing technology and connectivity, a guy sitting 5,000 miles away has better situational awareness than the guy in an aircraft.


___________________

Just Remember, We ALL Told You So.
 
Posts: 22442 | Location: Occupying Little Minds Rent Free | Registered: 04 October 2012Reply With Quote
one of us
Picture of f224
posted Hide Post
And a few years ago, they claimed the Indian Air Force pilots topped USAF F22's. But that was in a highly restrictive environment.

Everything I needed to know about UAV's I learned over Kandahar one night a few years ago. The local ATC controller asked if we had the "light civil" in sight. I responded yes we do. He replied that we might want to stay out of it's way, it had lost it's data link and was not responding as programmed. We watched it dive straight in from 13,000 feet...


Captain Dave Funk
Operator
www.BlaserPro.com
 
Posts: 822 | Location: Dallas, Iowa, USA | Registered: 05 June 2004Reply With Quote
one of us
Picture of MyNameIsEarl
posted Hide Post
Im not a pilot and never claimed to be but find it very interesting. Unless I read a different AI vs Pilot article, this wasn't necessarily a true test. During the sim, from I understand, the AI was given credit for kills anytime the nose of his aircraft passed the other pilot. Again, not taking into deflection or time on target.
 
Posts: 633 | Location: Camp Verde, AZ | Registered: 05 February 2006Reply With Quote
one of us
posted Hide Post
quote:
Originally posted by Todd Williams:
I'm not completely sure how to feel about this.

On the one hand, and obviously the most important aspect, anything that increases our combat effectiveness is a step in the right direction, reducing risk to our personnel and equipment.

On the other hand, as a former fighter pilot myself, I can't help but feel diminishment in the loss of the man to man contest of Ariel combat, often almost considered gentlemanly, although certainly a high stakes game with consequences of death or captivity very possible. That's the way Richthofen viewed it, and I think most of the guys I flew with held at least some semblance of respect for our opponents.


Aerial combat might be viewed as a waste of good pilots.


TomP

Our country, right or wrong. When right, to be kept right, when wrong to be put right.

Carl Schurz (1829 - 1906)
 
Posts: 12734 | Location: Moreno Valley CA USA | Registered: 20 November 2000Reply With Quote
one of us
posted Hide Post
Most fighter aircraft are limited by the G loading endured by the pilot, not the mechanical structure. The AI pilot doesn't suffer that same limitation.


John in Oregon
 
Posts: 620 | Location: Oregon | Registered: 23 November 2002Reply With Quote
one of us
Picture of Todd Williams
posted Hide Post
quote:
Originally posted by Johnly:
Most fighter aircraft are limited by the G loading endured by the pilot, not the mechanical structure. The AI pilot doesn't suffer that same limitation.


Not to be overly critical of your comment but what you are speaking of is the maneuvering aspect we fighter pilots refer to as Basic Fighter Maneuvering or BFM. It is quite different from Air Combat Maneuvering or ACM.

BFM is what most people think of as Dog Fighting and within that description, most don't understand the intricacies involved.

The focus on actual air to air combat is most definitely NOT BFM. It is ACM. The ACM focus is on exploiting tactical advantages of employing specific formations of multiple aircraft in order to maximize radar search / targeting effectiveness in combination with performance capabilities of the onboard weapons systems and existing Rules of Engagement.

In short, if, as a fighter pilot, you find yourself in the position of having to turn (engage in BFM) with the adversary (more likely adversaries [either a known or more likely, an unknown number of]), you are in serious trouble already. Much more important than G Load capabilities of the machine or man, is the ability to target an adversary with more acute angles off the aircraft's nose position and keep the weapon's guidance system focused on the target. The higher the angle off the nose you can launch and maintain targeting to impact, the further you extend your max engagement range while at the same time limiting your adversary's ability to effectively target you. If you, or your multi-ship formation, are not successful in eliminating the bogies prior to the merge, the best bet is to blow through, extend separation, and either pitch back for another run beyond visual range, or escape completely and regroup.

That's a very brief introduction to ACM considerations and a long way around the block to say that while G load capabilities are important if you find yourself in a last ditch effort air to air engagement (BFM), you are in serious trouble. There are many, many, much more important aspects to the air combat ball game.

All of this doesn't even scratch the surface of Rules of Engagement. In everything other than end of the world nuke war scenarios, Visual Identification (VID) will always be required. This requires a man in the machine. Electronic Identification (EID), which is what AI is capable of, doesn't satisfy typical ROE. Back to my mention of maximizing tactical advantages of specific formations, the VID ROE requirement is as much a part of that discussion as anything. A very good formation for this being the Battle Box with a flight of 4.

G force capabilities of a manned vs non-manned aircraft? A small consideration really.
 
Posts: 8107 | Registered: 09 January 2011Reply With Quote
one of us
Picture of richj
posted Hide Post
Wild ass Idea

USA AI beat USA Pilot. Is every pilot around the world trained to do the same thing for every situation?
 
Posts: 5360 | Location: NY, NY | Registered: 28 November 2005Reply With Quote
one of us
posted Hide Post
Read about John R Boyd, 40 sec Boyd.
 
Posts: 8854 | Location: Poetry, Texas | Registered: 28 November 2004Reply With Quote
  Powered by Social Strata  
 


Copyright December 1997-2021 Accuratereloading.com


Visit our on-line store for AR Memorabilia