A dogfight, or dog fight, is an aerial battle between fighter aircraft, conducted at close range. Modern terminology for air-to-air combat is air combat maneuvering (ACM), which refers to tactical situations requiring the use of individual basic fighter maneuvers (BFM) to attack or evade one or more opponents. Fighter pilots have to act fast when they’re in dogfights (close-range aerial battles), making split seconds decisions otherwise can cost them their lives. Pilots move at high speeds and need to avoid enemies while tracking them and keeping a contextual knowledge of objectives, terrains, fuel, and other key variables. Pilots need to rely on their creativity and decision-making abilities to survive.
Recently there was dogfight between India Pakistan airforce pilots on February 27, 2019, clearly impying that dogfight is still relevant. The Pakistan Air Force deployed “a large strike package” of modern F-16 Falcons, Chinese made JF-17s and some vintage Mirage-5 attack jets to avenge India’s bombing of terror sanctuaries in Balakot, Khyber-Pakhtunkhwa. The Indian Air Force scrambled six MiG-21s from its frontline air base in Srinagar to intercept the Pakistani fighters. In the ensuing dogfight, the first between India and Pakistan since the 1971 War, Wing Commander Varthaman in his Soviet-era jet managed to acquire a lock on one of the F-16s, shooting it down with a short-range Vympel R-73 air to air missile. Although he couldn’t see the outcome of the short 15-minute high-altitude dogfight, Varthaman radioed to base the words “R-73 selected”. Seconds later he was himself shot down.
DARPA has been devloping Artificial Intelligence (AI) based technology under the Air Combat Evolution (ACE) program to assist pilots in dogfighting by taking over low-level maneuvering tasks from fighter pilots. Earlier, Artificial intelligence was able to defeat chess grandmasters, Go champions, professional poker players, and, now, world-class human experts in the online strategy games Dota 2 and StarCraft II. However, No AI currently exists, that can outduel a human strapped into a fighter jet in a high-speed, high-G dogfight.
As modern warfare evolves to incorporate more human-machine teaming, DARPA seeked to automate air-to-air combat, enabling reaction times at machine speeds and freeing pilots to concentrate on the larger air battle.Turning aerial dogfighting over to AI is less about dogfighting, which should be rare in the future, and more about giving pilots the confidence that AI and automation can handle a high-end fight. To pursue this vision, DARPA created the Air Combat Evolution (ACE) program. ACE aims to increase warfighter trust in autonomous combat technology by using human-machine collaborative dogfighting as its initial challenge scenario.
In August 2020, the agency reported that an Artificial Intelligence has defeated Human Lockheed F-16 Pilot In Virtual Dogfight sponsored by it. The first round featured each team flying their algorithms vs. adversary AI algorithms. The teams then competed against each other in a round-robin style competition. After two days of competition, the winning algorithm of Darpa’s Air Combat Evolution program took on a human pilot in a Lockheed Martin (LMT) F-16 simulator. Artificial intelligence teams from Boeing (BA) subsidiary Aurora Flight Sciences, EpiSys Science, Georgia Tech Research Institute, Heron Systems, Lockheed Martin, Perspecta Labs, PhysicsAI, and SoarTech entered the competition. In a semifinal, Lockheed beat Physics AI. Heron defeated Aurora in the other semifinal and then took down Lockheed in the final. Heron scored five kills vs. zero for the human pilot.
In a future air domain contested by adversaries, a single human pilot can increase lethality by effectively orchestrating multiple autonomous unmanned platforms from within a manned aircraft. This shifts the human role from single platform operator to mission commander. In particular, 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.
ACE is one of several STO programs designed to enable DARPA’s “mosaic warfare” vision. Mosaic warfare shifts warfighting concepts away from a primary emphasis on highly capable manned systems — with their high costs and lengthy development timelines — to a mix of manned and less-expensive unmanned systems that can be rapidly developed, fielded, and upgraded with the latest technology to address changing threats. Linking together manned aircraft with significantly cheaper unmanned systems creates a “mosaic” where the individual “pieces” can easily be recomposed to create different effects or quickly replaced if destroyed, resulting in a more resilient warfighting capability.
Linking together manned aircraft with significantly cheaper unmanned systems creates a “mosaic” where the individual “pieces” can easily be recomposed to create different effects or quickly replaced if destroyed, resulting in a more resilient warfighting capability.The ACE program will train AI in the rules of aerial dogfighting similar to how new fighter pilots are taught, starting with basic fighter maneuvers in simple, one-on-one scenarios.
DARPA’s Air Combat Evolution (ACE) program
The key to victory is to make more appropriate decisions more quickly than one’s opponents. This critical insight, developed by U.S. Air Force Colonel John Boyd, is embodied in his OODA Loop concept. The OODA Loop is an iterative model for decision-making consisting of 4 phases: observation, orientation, decision-making, and action.Importantly, it is a loop whereby the outcomes of actions may be observed and influence subsequent decision-making.
As modern warfare evolves to incorporate more human-machine teaming, DARPA seeks to automate air-to-air combat, enabling reaction times at machine speeds and freeing pilots to concentrate on the larger air battle. DARPA wants to built artificial intelligence systems to follow the OODA Loop , making intelligent systems that make even faster more accurate decisions. Recent advances in AI suggest that we are on the cusp of such a revolution. Deep learning has had particular success in the “observe” part of the OODA Loop. Recent advances are yielding remarkable understanding of images, video, audio, speech, and text, all of which are perceptual tasks.
Great progress is being made in decision-making using technologies like evolutionary computation (EC) and Bayesian optimization. Meanwhile, we are seeing slower progress with regard to reasoning or “orientation”. However, we are not yet fully exploiting many advances in large-scale graphical models, which present new opportunities in the context of the OODA Loop,writes Nigel Duffy, Global Innovation AI Leader, Ernst & Young LLP. Particularly compelling are the developments in real-time black-box optimization. Black-box optimization is an ideal approach to the problem of achieving an optimal outcome under uncertainty —where one must both make good decisions and learn how to make good decisions at the same time. By iterating and learning quickly, the quality of such decisions becomes better and better.
As soon as new human fighter pilots learn to take-off, navigate, and land, they are taught aerial combat maneuvers. Contrary to popular belief, new fighter pilots learn to dogfight because it represents a crucible where pilot performance and trust can be refined. To accelerate the transformation of pilots from aircraft operators to mission battle commanders — who can entrust dynamic air combat tasks to unmanned, semi-autonomous airborne assets from the cockpit — the AI must first prove it can handle the basics.
The end goal is to have autonomous jet controls that can handle tasks like dodging out the way of enemy fire at lightning speeds, while the pilot takes on more difficult problems like executing strategic battle commands and firing off weapons.“We envision a future in which AI handles the split-second maneuvering during within-visual-range dogfights, keeping pilots safer and more effective as they orchestrate large numbers of unmanned systems into a web of overwhelming combat effects,” said Lieutenant Colonel Dan Javorsek, ACE program manager.
Turning aerial dogfighting over to AI is less about dogfighting, which should be rare in the future, and more about giving pilots the confidence that AI and automation can handle a high-end fight. As soon as new human fighter pilots learn to take-off, navigate, and land, they are taught aerial combat maneuvers. Contrary to popular belief, new fighter pilots learn to dogfight because it represents a crucible where pilot performance and trust can be refined. To accelerate the transformation of pilots from aircraft operators to mission battle commanders — who can entrust dynamic air combat tasks to unmanned, semi-autonomous airborne assets from the cockpit — the AI must first prove it can handle the basics.
The ACE program seeks to increase trust in combat autonomy by using human-machine collaborative dogfighting as its challenge problem, which also serves as an entry point into developing complex human-machine teaming. The program is implementing methods to predict, measure, calibrate, and increase human trust in the autonomy’s performance. Additionally, ACE will scale the tactical application of autonomous dogfighting to more complex, heterogeneous, multi-aircraft, operational-level simulated scenarios. These scenarios will be informed by live data, laying the groundwork for future live, campaign-level Mosaic Warfare experimentation.
“While much of the excitement surrounding ACE is based upon the success of AlphaDogfight Trials and the promise of autonomous tactical air combat, the program is ultimately focused on developing a protocol for teaching humans to trust autonomy and to develop more advanced human-machine symbiosis,” said Tim Grayson, director of DARPA’s Strategic Technology Office. “The award of these TA1 contracts represents the first step toward developing the AI side of that partnership.”
Center for New American Security’s Paul Scharre told the BBC March, “Swarming allows you to build large numbers of low-cost expendable agents that can be used to overwhelm an adversary,” adding “And unlike having a large number of soldiers, robotic agents can coordinate on a scale that would be impossible for humans.” But before controlling a swarm of drones, human pilots need to start working with a single or couple of them to begin with.
ACE creates a hierarchical framework for autonomy in which higher-level cognitive functions (e.g., developing an overall engagement strategy, selecting and prioritizing targets, determining best weapon or effect, etc.) may be performed by a human, while lower-level functions (i.e., details of aircraft maneuver and engagement tactics) is left to the autonomous system. In order for this to be possible, the pilot must be able to trust the autonomy to conduct complex combat behaviors in scenarios such as the within visual range dogfight before progressing to beyond visual range engagements.
The ACE demonstrations will bridge the gap from simple physics-based automated systems currently in use to complex systems capable of effective autonomy within highly dynamic and uncertain environments at mission speeds.
The technology development on the ACE program addresses four primary challenges:
- Increase air combat autonomy performance in local behaviors (individual aircraft and team tactical)
- Build and calibrate trust in air combat local behaviors
- Scale performance and trust to global behaviors (heterogeneous multi-aircraft)
- Build infrastructure for full-scale air combat experimentation
DARPA held a Proposers Day for interested researchers on May 17, 2019, in Arlington, Virginia.“Being able to trust autonomy is critical as we move toward a future of warfare involving manned platforms fighting alongside unmanned systems,” said Air Force Lt. Col. Dan Javorsek (Ph.D.), ACE program manager in DARPA’s Strategic Technology Office (STO). “We envision a future in which AI handles the split-second maneuvering during within-visual-range dogfights, keeping pilots safer and more effective as they orchestrate large numbers of unmanned systems into a web of overwhelming combat effects.”
While highly nonlinear in behavior, dogfights have a clearly defined objective, measureable outcome, and the inherent physical limitations of aircraft dynamics, making them a good test case for advanced tactical automation. Like human pilot combat training, the AI performance expansion will be closely monitored by fighter instructor pilots in the autonomous aircraft, which will help co-evolve tactics with the technology.
These subject matter experts will play a key role throughout the program.“Only after human pilots are confident that the AI algorithms are trustworthy in handling bounded, transparent and predictable behaviors will the aerial engagement scenarios increase in difficulty and realism,” Javorsek said. “Following virtual testing, we plan to demonstrate the dogfighting algorithms on sub-scale aircraft leading ultimately to live, full-scale manned-unmanned team dogfighting with operationally representative aircraft.”
The ACE program will train AI in the rules of aerial dogfighting similar to how new fighter pilots are taught, starting with basic fighter maneuvers in simple, one-on-one scenarios. While highly nonlinear in behavior, dogfights have a clearly defined objective, measureable outcome, and the inherent physical limitations of aircraft dynamics, making them a good test case for advanced tactical automation. Like human pilot combat training, the AI performance expansion will be closely monitored by fighter instructor pilots in the autonomous aircraft, which will help co-evolve tactics with the technology. These subject matter experts will play a key role throughout the program.
“Only after human pilots are confident that the AI algorithms are trustworthy in handling bounded, transparent and predictable behaviors will the aerial engagement scenarios increase in difficulty and realism,” Javorsek said. “Following virtual testing, we plan to demonstrate the dogfighting algorithms on sub-scale aircraft leading ultimately to live, full-scale manned-unmanned team dogfighting with operationally representative aircraft.”
DARPA seeks a broad spectrum of potential proposers for each area of study, including small companies and academics with little previous experience with the Defense Department. To that end, before Phase 1 of the program begins, DARPA will sponsor a stand-alone, limited-scope effort focused on the first technical area: automating individual tactical behavior for one-on-one dogfights.
Called the “AlphaDogfight Trials,” this initial solicitation will be issued by AFWERX, an Air Force innovation catalyst with the mission of finding novel solutions to Air Force challenges at startup speed. The AFWERX trials will pit AI dogfighting algorithms against each other in a tournament-style competition.“Through the AFWERX trials, we intend to tap the top algorithm developers in the air combat simulation and gaming communities,” Javorsek said. “We want them to help lay the foundational AI elements for dogfights, on which we can build as the program progresses.”
Dynetics awarded DARPA Air Combat Evolution (ACE) Phase 1 in May 2020
Dynetics, Inc., a wholly owned subsidiary of Leidos (NYSE:LDOS), has been awarded Phase 1 of the Air Combat Evolution (ACE) program, Technical Area 3 (TA3), by the Defense Advanced Research Projects Agency’s (DARPA) Strategic Technology Office (STO). ACE TA3, also known as Alpha Mosaic, is valued at $12.3 million. The ACE program is using aerial dogfighting as the initial challenge scenario for implementing artificial intelligence (AI) into high-intensity air conflicts, which intends to increase warfighter trust in combat autonomy. Similar to how the United States military trains fighter pilots, ACE performers will work to increase trust in automated, within-visual-range, air-to-air dogfighting. As algorithms and tactics mature, so will the scenarios and adversarial capabilities.
During the 18-month Phase 1 award, Dynetics will leverage program advances in automated dogfighting to enable operational-level scenarios with a large number of heterogeneous aircraft. By improving the algorithms and tactics developed within ACE, TA3 will lay the groundwork for future live, campaign-level experimentation of manned and unmanned vehicles. “The ACE program is inspiring on so many levels,” said Tim Keeter, ACE program manager for Dynetics. “Our team brings novel solutions that have proven to be feasible and scalable to these challenging ACE objectives. These efforts will help DARPA and the U.S. military expand their advantage in the evolution of Mosaic warfare.”
The program consists of three phases. Phase 1 begins research in a simulated environment. Phase 2 advances to a flight environment using unmanned air vehicles. Phase 3 includes a realistic, manned-flight environment involving complex human-machine collaboration. “Our entry into Phase 1 of ACE represents years of relevant research within Dynetics and our team members that position us to do great things for our country,” said Kevin Albarado, Dynetics’ chief engineer. “Our scientists and engineers are eager to continue advancing these state-of-the-art AI applications to help our warfighters defend our nation.” Dynetics has formed an industry team representing best-in-class leadership to tackle these technological challenges, they include: Soar Technology, Inc. (SoarTech), InfoSciTex, and Intuitive Research & Technology Corporation (IRTC). The Dynetics team will leverage its experience of advancing second and third wave artificial intelligence and autonomy concepts. Many of these concepts are performed within the scope of other major DARPA programs.
DARPA Awards Calspan Air Combat Evolution contract in July 2020
Calspan Corporation has been awarded a $14.1M, four-year contract by the Defense Advanced Research Projects Agency (DARPA) to develop full-scale air combat experimentation infrastructure for its Air Combat Evolution (ACE) program. Under this contract Calspan Flight Research will modify up to four Aero Vodochody L-39 Albatros jet trainers with Calspan’s proprietary autonomous fly-by-wire flight control system technology to allow implementation and demonstration of advanced Human Machine Interfaces (HMI) and AI algorithms. Flight tests and demonstrations will be conducted from the Calspan Flight Research Facility at the Niagara Falls, NY, International Airport and flown in the Misty Military Operating Area (MOA) over nearby Lake Ontario. The program will be conducted over three phases of development with Phase 1 beginning immediately.
For more than 75 years, Calspan has been providing research and testing services in the aviation and transportation industries. Internationally recognized for safety research and innovation, the company’s headquarters is located in Buffalo, NY, housing a variety of research and testing facilities, including a transonic wind tunnel, ground vehicle crash testing, dynamic sled testing and research, and tire research and performance testing. Calspan conducts flight testing, flight training, and aircraft modifications in Niagara Falls, NY where they own a fleet of seven airborne testbeds. Calspan also has prototype test device manufacturing capabilities in Newport News, Virginia, as well as force measurement equipment manufacturing capabilities in San Diego, California.
DARPA Selects Teams to Further Advance Dogfighting Algorithms in Nov 2020
DARPA awarded contracts to five companies to develop algorithms enabling mixed teams of manned and unmanned combat aircraft to conduct aerial dogfighting autonomously. Boeing, EpiSci, Georgia Tech Research Institute, Heron Systems, and physicsAI were chosen to develop air combat maneuvering algorithms for individual and team tactical behaviors under Technical Area (TA) 1 of DARPA’s Air Combat Evolution (ACE) program. Each team is tasked with developing artificial intelligence agents that expand one-on-one engagements to two-on-one and two-on-two within-visual-range aerial battles. The companies’ algorithms will be tested in each of three program phases: modeling and simulation, sub-scale unmanned aircraft, and full-scale combat representative aircraft scheduled in 2023.
“The TA1 performers include a large defense contractor, a university research institute, and boutique AI firms, who will build upon the first-gen autonomous dogfighting algorithms demonstrated in the AlphaDogfight Trials this past August,” said Air Force Col. Dan “Animal” Javorsek, program manager in DARPA’s Strategic Technology Office. “We will be evaluating how well each performer is able to advance their algorithms to handle individual and team tactical aircraft behaviors, in addition to how well they are able to scale the capability from a local within-visual-range environment to the broader, more complex battlespace.”
With the selection of the TA1 algorithm developers, performers for the program’s four technical areas are all now on contract. Performers for TAs 2-4 were selected earlier this year. TA2 performer SoarTech is developing experimental methodology for modeling and measuring pilot trust in dogfighting autonomy as well as novel human-machine interfaces (HMI). TA3 performers Dynetics and Lockheed Martin are developing a data set and model for large force exercise data analytics to evaluate how well TA1 algorithms scale to larger, more complex multi-aircraft scenarios. TA4 performer Calspan is supplying full-scale L-39 aircraft for Phase 3 of the program, with the goal of demonstrating TA1 autonomous dogfighting algorithms and TA2 HMIs on full-scale combat aircraft with a human safety pilot on board.
DARPA Trialled AI-Based Dogfight Technology
The first two trials took place in November 2019 and January 2020 near Baltimore at the Johns Hopkins University Applied Physics Laboratory, which is developing and managing the simulation air environment for the AlphaDogfight Trials. Eight teams were selected to compete in the AlphaDogfight trials, a virtual competition designed to demonstrate advanced artificial intelligence (AI) algorithms that can perform simulated within-visual-range air combat maneuvering, colloquially known as a dogfight.
The Trials aim to energize and expand a base of AI developers and potential proposers prior to an anticipated algorithm-development solicitation to be released under DARPA’s Air Combat Evolution (ACE) program. Announced earlier this year, ACE seeks to automate air-to-air combat with trusted AI able to manage lower-order operations, pilots could focus on higher-order strategic challenges. “Through the AFWERX trials, we intend to tap the top algorithm developers in the air combat simulation and gaming communities,” Javorsek said. “We want them to help lay the foundational AI elements for dogfights, on which we can build as the program progresses.”
F-16 jet to boost trust in artificial intelligence (AI) reported in Feb 2022
Officials of the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va., have issued a solicitation (HR001122S0015) for the Air Combat Evolution (ACE) Full-Scale Aircraft TA-4 project, which seeks to increase trust in combat autonomy using human-machine collaboration in aircraft dogfighting.
After a successful first phase, the ACE program has entered its second phase. This solicitation asks industry for proposals to convert existing F-16 aircraft into human-in-the-loop, safety-sandboxed testbed aircraft to support autonomy development and experimentation.
The solicitation involves technology areas that call for additional aircraft hardware and additional aircraft mission systems software integration to support autonomous within-visual-range maneuvering and trust research in the ACE program.
The additional aircraft options will support ACE as well as a wider range of autonomy development needs. The ACE project also will develop enabling technologies to enhance collaboration among humans and unmanned combat aircraft in a variety of combat scenarios.
The program is scaling machine automation in aircraft dogfighting to more complex, heterogeneous, multi-aircraft, operational level simulated scenarios informed by live data. These scenarios are expected to lay the groundwork for future live, campaign-level experiments.
The idea is to enable one human pilot to become a more deadly warfighter by leading several semi-autonomous artificially intelligent unmanned aircraft, all from his own cockpit. This would shift the human role from sole operator to system mission commander.
References and resources also include: