9 mai 2019 | International, Aérospatial, Autre défense

DARPA: Training AI to Win a Dogfight

Artificial intelligence has defeated chess grandmasters, Go champions, professional poker players, and, now, world-class human experts in the online strategy games Dota 2 and StarCraft II. No AI currently exists, however, 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 seeks 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. 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.

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. DARPA will hold 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.”

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.

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.”

AFWERX will announce the trials in the near future on its website: https://www.afwerx.af.mil/.

For ACE Proposers Day registration details, please visit: https://go.usa.gov/xmnMn

https://www.darpa.mil/news-events/2019-05-08

Sur le même sujet

  • SASC chairman: We must build the national security innovation base our defense strategy requires

    2 décembre 2019 | International, Aérospatial, Naval, Terrestre, C4ISR, Sécurité

    SASC chairman: We must build the national security innovation base our defense strategy requires

    By: Sen. Jim Inhofe Since World War II, the American people have believed our military has had the best of everything, but the technological superiority that kept us 20 years ahead of our competitors has rapidly diminished. In some cases, we're already behind. By 2030, unless we pursue “urgent change at significant scale,” as former Defense Secretary Gen. James Mattis put it, it's likely the U.S. will face an enemy with superior weapons, superior equipment and superior capabilities. Nowhere is this better illustrated than in our strategic competition with China. China used to just steal our technology. Now, through heavy investment, they are improving it. The result? China is outpacing the U.S. in key areas like hypersonic weapons, artificial intelligence and biotechnologies — not to mention conventional capabilities. China isn't the only one. Technological development is accelerating across the globe, expanding to more actors and changing the very nature of war. We can't afford to let our advantage erode further. It is up to the Department of Defense and Congress to make sure that the defense-industrial base becomes, as the National Defense Strategy demands, an “unmatched 21st century National Security Innovation Base.” If we want to “sustain security and solvency,” we need to consider wholesale change to industry culture and its interface with the Department of Defense, shed outdated management processes, and reimagine a resilient supply chain that mitigates 21st century risks. This essay is part of the Defense News 2020 Outlook project. Click here for more. This begins with software, which is foundational to military capability. The DoD and its traditional hardware-dominant industry partners have been behind on software in almost every way — talent, tools, development and delivery processes. Software innovation has failed in countless DoD programs, including the Ford-class carrier, the F-35′s Autonomic Logistics Information System and the GPS next-generation operational control system. Instead of taking the Pentagon for granted as an endless source of cash flow, partners must refocus their attention on delivering secure capability that actually works. Next, the Department of Defense needs to continue to expand capacity — prioritizing speed of delivery and adapting its systems to maximize value and output. For too long we have been slow to expand our stockpiles of fifth-generation weapons required to fight peer adversaries. The second production line for JASSM-ER cruise missiles is a good start toward building the capacity needed to retain advantages that will make any enemy think twice before attacking. We must do the same for other fifth-generation weapons, including air-to-air missiles. Shipbuilding, including aircraft carriers, surface ships, submarines and our logistics fleet, is another area where our capacity is severely limited. The Chinese People's Liberation Army Navy, which recently surpassed ours in size, is on track to reach 400 ships in 2025 and is nearly self-sufficient for all components. Size of the fleet isn't a sole consideration. We've focused on ensuring the capability of our fleet remains unmatched and bolstering suppliers of critical components, but we must also improve the construction performance of lead ships in new classes to maintain and build upon our capability advantage. The last thing we want is a fair fight. Innovation is best done at the subsystem level through a rigorous engineering-based process centered on building knowledge through full-scale prototypes, which can then inform ship design. We are eager to work with the Navy to identify and fund more of these prototypes, which will serve as the building blocks of the future fleet. We also must accelerate innovation. Recent defense authorization legislation encourages the DoD to streamline acquisition, take a business-minded approach to contracting, and tap into nontraditional suppliers and public-private partnerships. This must continue. Dilapidated testing infrastructure is holding us back from catching up to our enemies. Just look at hypersonic weapons: Beijing is parading around dozens of its newest weapons, and we have yet to build one. The DoD has looked to Silicon Valley, but we are competing with Chinese influence there as well, and the Pentagon has often proven an impossible customer due to its antiquated bureaucracy. Any technological improvements will be meaningless if vulnerable to being infiltrated or stolen. Recent legislation continues support for the DoD as it assesses and mitigates risks to its supply chains posed by adversaries. Both the government and contractors need to cooperate on and use modern verification tools to identify trusted suppliers and manufacturers, as well as fix vulnerabilities. To make these tools useful, the DoD must first establish a working digital model of its suppliers. Lastly, while we must continue to invest in the domestic, organic industrial base, it's important to remember that we can't take on China and Russia alone — which is why the National Defense Strategy emphasizes our network of allies and partners. We must remove unnecessary barriers to industrial cooperation that degrade our collective competitive edge. We do not have to make a false choice between investing domestically and in our allies — we can do both. Under our National Technology and Industrial Base partnership with Canada, Australia and the United Kingdom, we can develop a more diverse, resilient industrial base, secure our supply chains, and become a “five eyes for defense procurement.” It's in our best interest to ensure our allies can leverage our technological advantages and we can leverage theirs. Without a strong national security innovation base, the Pentagon cannot implement the National Defense Strategy. Congress' job is to put the appropriate, tailored policy in place and provide sufficient, predictable resources to help the industrial base meet these challenges. Together, we can harness the power of American innovation to ensure that we are able to win the wars of the future. https://www.defensenews.com/outlook/2019/12/02/sasc-chairman-we-must-build-the-national-security-innovation-base-our-defense-strategy-requires

  • Does DoD know how to supply intel for cyber ops?

    3 juillet 2018 | International, C4ISR

    Does DoD know how to supply intel for cyber ops?

    By: Mark Pomerleau Cyber has been an official domain of warfare for nearly a decade, yet the Department of Defense is still learning how to integrate it with operations. And some members of Congress are concerned the traditional military intelligence organs to this day don't understand intel support to cyber ops. The House Armed Services Committee is directing that a briefing on the subject must take place by December 1, 2018. The briefing — delivered by the under secretary of defense for intelligence, in coordination with the Defense Intelligence Agency and the military services — is expected, according to a provision in the committee's annual defense policy bill, to address multiple issues, including: Efforts to standardize a common military doctrine for intelligence preparation of the battlefield for cyber operations; Efforts to develop all-source intelligence analysts with the capability to support cyber operations; and Efforts to resource intelligence analysis support elements at U.S. Cyber Command and the service cyber components. “The committee is concerned about the Defense Intelligence Enterprise's ability to provide the cyber community with all-source intelligence support, consistent with the support provided to operations in other domains,” the provision, called an “item of special interest,” says. In some cases, other intelligence disciplines, such as human intelligence or signals intelligence, might be needed to help enable a cyber operation. A committee aide noted that the goal is to get DoD to think about cyber operations just as operations in any domain and build the infrastructure to support that. According to Gus Hunt, Accenture Federal Services cyber strategy lead, cyber as a domain is really no different than the others from an intelligence support perspective. The objective of intelligence, he told Fifth Domain in a recent interview, is to ensure it provides timely information about the adversary, who they are, the status of their capabilities and any information about the threats that are there. “I think what you're seeing ... is that people are asking the question are we appropriately structured or resourced and focused to be as effective as we possibly can in this new realm of cyber and cyber operations,” Hunt, who previously served as the chief technology officer at the CIA, said. “Because they're asking the question, I think the obvious answer is ... we're not structured as effectively as we possibly can be ... [but] it's really good that people are sitting there asking.” The Army is experiencing similar problems, especially when it comes to experimenting with force structure changes and bringing cyber effects to the tactical edge, which currently don't exist. “We're not seeing a corresponding growth in the intel organizational structure with the cyber and” electronic warfare, Lt. Col. Chris Walls, deputy division chief for strategy and policy in the cyber directorate of the Department of the Army G-3/5/7, said at the C4ISRNET conference in May. “The existing intel force structure is really going to be stressed when we put this EW and cyber capability into the field unless they have a corresponding growth and capability as well,” Walls said of tactical cyber effects and teams. https://www.fifthdomain.com/congress/2018/07/02/does-dod-know-how-to-supply-intel-for-cyber-ops/

  • Lockheed, U.S. government offer to sell 40 F-35As to Swiss air force

    20 novembre 2020 | International, Aérospatial

    Lockheed, U.S. government offer to sell 40 F-35As to Swiss air force

    By Christen McCurdy Nov. 19 (UPI) -- The U.S. government and Lockheed Martin submitted a proposal to the Swiss government offering up to 40 F-35A aircraft and a sustainment and training program. According to Lockheed Martin, the offering also includes an industrial package that would provide Swiss industry with work opportunities in all regions of the country. If the deal goes through, Lockheed said, Swiss industry would have the opportunity to compete for direct production of F-35 components as well as sustainment projects to support the Swiss air force and Swiss autonomy and cybersecurity projects related to the F-35. The offer also includes a six-month sustainment package that would allow the Swiss air force to conduct autonomous operations. The option would permit the assembly of four aircraft in Switzerland to help the Swiss air force and industry partners learn how to maintain the aircraft. "We are confident that our F-35 offer is the best and most affordable solution for the Swiss NFA competition," said Greg Ulmer, F-35 Program vice president and general manager. "We are offering the only 5th generation fighter at the cost of 4th generation aircraft while offering Switzerland an aircraft that will protect Swiss sovereignty for decades to come." Nine nations operate F-35s on their home soil, with more than 585 F-35s in service today, according to Lockheed. https://www.upi.com/Defense-News/2020/11/19/Lockheed-US-government-offer-to-sell-40-F-35As-to-Swiss-air-force/3561605814292/

Toutes les nouvelles