Back to news

January 4, 2019 | International, C4ISR

DARPA: Generating Actionable Understanding of Real-World Phenomena with AI

DARPA seeks to develop schema-based AI capability to enhance reasoning about complex world events and generate actionable insights

Rapid comprehension of world events is critical to informing national security efforts. These noteworthy changes in the natural world or human society can create significant impact on their own, or may form part of a causal chain that produces broader impact. Many events are not simple occurrences but complex phenomena composed of a web of numerous subsidiary elements – from actors to timelines. The growing volume of unstructured, multimedia information available, however, hampers uncovering and understanding these events and their underlying elements.

“The process of uncovering relevant connections across mountains of information and the static elements that they underlie requires temporal information and event patterns, which can be difficult to capture at scale with currently available tools and systems,” said Dr. Boyan Onyshkevych, a program manager in DARPA's Information Innovation Office (I2O).

The use of schemas to help draw correlations across information isn't a new concept. First defined by cognitive scientist Jean Piaget in 1923, schemas are units of knowledge that humans reference to make sense of events by organizing them into commonly occurring narrative structures. For example, a trip to the grocery store typically involves a purchase transaction schema, which is defined by a set of actions (payment), roles (buyer, seller), and temporal constraints (items are scanned and then payment is exchanged).

To help uncover complex events found in multimedia information and bring them to the attention of system users, DARPA created the Knowledge-directed Artificial Intelligence Reasoning Over Schemas (KAIROS) program. KAIROS seeks to create a schema-based AI capability to enable contextual and temporal reasoning about complex real-world events in order to generate actionable understanding of these events and predict how they will unfold. The program aims to develop a semi-automated system capable of identifying and drawing correlations between seemingly unrelated events or data, helping to inform or create broad narratives about the world around us.

KAIROS' research objectives will be approached in two stages. The first stage will focus on creating schemas from large volumes of data by detecting, classifying and clustering sub-events based on linguistic inference and common sense reasoning. Researchers taking on this challenge will apply generalization, composition and specialization processes to help generate schemas that describe both simple and complex events, sequence multiple schemas together to understand key contextual elements like roles and timelines, and apply domain-specific knowledge to tailor the analysis for a particular need.

The second stage of the program will focus on applying the library of schemas created during stage one to multimedia, multi-lingual information to uncover and extract complex events. This stage will require identifying events and entities, as well as relationships among them to help construct and extend a knowledge base.

DARPA will hold a Proposers Day on January 9, 2019 from 10:00am to 2:30pm (EST) at the Holiday Inn at Ballston, 4610 N. Fairfax Drive, Arlington, Virginia 22203 to provide more information about KAIROS and answer questions from potential proposers. For details of the event, including registration requirements, visit https://www.schafertmd.com/darpa/i2o/KAIROS/pd/.

A Broad Agency Announcement that fully describes the program structure and objectives can be found here, https://www.fbo.gov/index?s=opportunity&mode=form&id=0fc6d1a237556c5d59847e7165af3aef&tab=core&_cview=1.

Image Caption: This image outlines the two stages of the KAIROS program. The first stage will focus on creating a library of schemas from large volumes of data by detecting, classifying and clustering sub-events based on linguistic inference and common sense reasoning. The second stage will apply those schemas to new information to uncover and extract complex events, as well as relationships among them, to help construct and extend a knowledge base.

https://www.darpa.mil/news-events/2019-01-04

On the same subject

  • Rafael finds European partners to market Trophy active protection system

    November 15, 2021 | International, Land

    Rafael finds European partners to market Trophy active protection system

    The new Germany-based venture, dubbed EuroTrophy, is charged with finding new takers for the defensive technology and leading any vehicle-integration efforts for future customers.

  • For US Army’s future command posts, one size will not fit all

    October 8, 2023 | International, Land

    For US Army’s future command posts, one size will not fit all

    “Each of these commanders is going to want to tailor their command post,” said Mark Kitz, the leader of the Army's PEO C3T.

  • OMFV: Army Seeks Industry Advice On Bradley Replacement

    February 27, 2020 | International, Land

    OMFV: Army Seeks Industry Advice On Bradley Replacement

    Having rebooted the Optionally Manned Fighting Vehicle program, the Army is now is asking industry input on how to achieve nine goals, from survivability to mobility to streamlined logistics. By SYDNEY J. FREEDBERG JR.on February 26, 2020 at 4:01 AM Two months ago, the Army cancelled its original solicitation to replace the M2 Bradley troop carrier after no company could meet the strict requirements. This afternoon, the Army officially asked for industry input on how to achieve nine broadly-defined “characteristics” for the future Optionally Manned Fighting Vehicle. “Feedback may be submitted in any form (concepts, information papers, technical papers, sketches, etc.),” says the announcement on SAM.gov. “The Army would like to obtain this initial feedback prior to 06 March 2020.” This call for suggestions on how to move forward comes just weeks after the Army issued a surprisingly apologetic survey asking industry what they did wrong the first time around. It's part of a newly humble approach in which the Army doesn't prescribe formal requirements up-front but instead lays out broad objectives and asks industry how best to achieve them. The chief of Army Futures Command, Gen. Mike Murray, gave reporters a preview of the nine characteristics three weeks ago, but the list announced today is much more detailed – though still leaving plenty of room for companies to brainstorm solutions. Our annotated highlights from the announcement – the emphasis is in the original: Background: The OMFV, as part of an Armored Brigade Combat Team (ABCT), will replace the Bradley to provide the capabilities required to defeat a future near-peer competitor's force. The Army is seeking a transformational increase in warfighting capability, not simply another incremental improvement over the current Bradley Fighting Vehicle. Concept of employment: As part of an ABCT, the OMFV will not fight alone, but rather as part of a section, platoon, and company of mechanized infantry.... “Near-peer competitor” is Pentagon jargon for “China or Russia” – chiefly Russia in this case, since the plains of Eastern Europe are a far more likely arena for armored warfare than Pacific islands. That the Army wants “transformational” improvements, not “incremental” ones, shows there's still some real ambition in the vision for this vehicle. At the same time, the OMFV will still fight “as part of an ABCT,” meaning the existing Armored Brigade Combat Team organization — not as part of some all-new organization with all-new equipment, as was once envisioned for the cancelled Future Combat Systems. Survivability. The OMFV must protect the crew and Soldiers from emerging threats and CBRN environments. The OMFV should reduce likelihood of detection by minimizing thermal, visual, and acoustic signatures. In other words, the vehicle needs to give the crew a chance of survival against cutting-edge anti-tank missiles, precision-guided artillery, attack drones and other such “emerging threats,” as chemical, biological, radiological, and nuclear contamination (CBRN). That does not mean the vehicle itself has to survive intact. The way this is worded, if a hit totals the OMFV but the soldiers inside can walk away, the Army will count that as a win. (The JLTV 4×4 truck takes this same approach to roadside bombs). So the OMFV doesn't necessarily have to have heavy armor protecting the entire vehicle. It could have a heavily armored crew compartment, light armor elsewhere, and an Active Protection System to intercept incoming threats. (The Russian T-14 Armata uses this combination). It also should avoid being spotted in the first place by eye, ear, or thermal sensor, which might favor designs with hybrid-electric motors that can switch from hot, noisy diesels to a battery-driven stealth mode. Mobility. The OMFV must have mobility that can keep pace with the Abrams in a combined arms fight through rural and urban terrain. That's the M1 Abrams main battle tank, which the existing M2 Bradley and M109 Paladin howitzer were also designed to keep up with. This is another aspect of that “concept of employment” that calls for the OMFV to slot into existing formations and work closely with existing vehicles. Note also the reference to “rural and urban terrain,” which will come up again: Traditionally the Army has avoided city fighting, but as urban sprawl covers ever more of the planet, technology and tactics have to adapt to brutal close-quarters combat. Growth. The OMFV must possess the growth margins and open architecture required for rapid upgrades and insertion of future technologies such as mission command systems, protection systems, and sensors. This characteristic is really where you get the potential for “transformational” improvements. The M2 Bradley was originally introduced in 1980 and, after 40 years of upgrades, it has very little margin left to handle additional weight or – even more important nowadays – power-hungry electronics. The Bradley's lack of room to grow has driven the Army to try replacing it three times already: the original OMFV requirements cancelled this year; the Ground Combat Vehicle cancelled in 2014; and the Future Combat Systems cancelled in 2009. Hopefully, fourth time's the charm. Lethality. The OMFV-equipped platoons must defeat future near-peer soldiers, infantry fighting vehicles, helicopters, small unmanned aerial systems, and tanks as part of a Combined Arms Team in rural and urban terrain. This is a more ambitious hit list than the Bradley, which sports machineguns for killing infantry, a 25 mm autocannon to destroy light armored vehicles, and the obsolescent TOW missile for taking on heavy tanks. The Pentagon is increasingly worried about small drones, which ISIS terrorists have used as flying IEDs and Russian artillery has used as spotters for barrages. With Russia and China developing increasingly sophisticated anti-aircraft systems, there's also a concern that US fighters may not be able to keep enemy attack helicopters at bay, forcing ground forces to handle that threat themselves. These aerial targets require more sophisticated tracking systems, and drones may be best dealt with by electronic jamming or lasers rather than bullets. Weight. The OMFV must traverse 80% of Main Supply Routes (MSRs), national highways, and bridges in pacing threat countries, and reduce the cost of logistics and maintenance. Designs must allow for future growth in components and component weights without overall growth of vehicle weight through modularity and innovation. Weight is the issue that has bedeviled Bradley replacements for two decades. The FCS vehicles, optimized for air transport, were too light to carry adequate armor; GCV was too heavy; and the original OMFV couldn't meet its air transport requirements and its protection requirements at the same time. With most bridges in Eastern Europe unable to safely take weights over 50 tons, too much heavy armor can cripple your mobility. Logistics. The OMFV must reduce the logistical burden on ABCTs and must be equipped with advanced diagnostic and prognostic capabilities. Advanced manufacturing and other innovative techniques should be included in the design that reduce the time and cost of vehicle repairs. There are two big factors that make a vehicle hard to keep supplied and in working order. One is weight – heavier vehicles burn more fuel – and the other is complexity. High-tech is usually high-maintenance. The US military is hopeful that AI-driven predictive maintenance can detect and head off impending breakdowns, and that 3D printing can produce at least some spare parts on demand without a long supply line. Transportability. The OMFV must be worldwide deployable by standard inter- and intra-theater sea, waterway, air, rail, and road modes of transportation. The original OMFV requirement very specifically called for two of the vehicles to fit on a single Air Force C-17 jet transport, which proved undoable with the weight of armor desired. This time, the Army isn't specifying any particular aircraft. In practice, armored vehicles are almost always shipped by sea and, where possible, stockpiled on allied soil well before a crisis erupts. On land, since tracked vehicles aren't designed to drive hundreds of miles by road, they're usually deployed to the battle zone by train or tractor-trailer, both of which have their own weight limits. Manning. The OMFV should operate with the minimal number of crew members required to fight and win. The OMFV should allow commanders to choose between manned or remote operation based on the tactical situation. This is the objective that gave the OMFV its name: Optionally Manned Fighting Vehicle. Now, since it's a Bradley replacement, the OMFV is supposed to be a troop carrier – specifically, the heavily armed and armored kind known as an Infantry Fighting Vehicle – so by definition it needs to carry people. But the Army is intensely interested in having the option to run it by remote control, or maybe even autonomously, to (for example) scout out especially dangerous areas or carry casualties back to an aid post without pulling healthy soldiers out of the fighting line. Training. The OMFV should contain embedded training capabilities that are compatible with the Synthetic Training Environment (STE). STE is the Army's total overhaul of its training simulators, drawing on commercial gaming technology to develop an array of virtual and augmented reality systems using a common database of real-world terrain. Instead of having to use a simulator in a warehouse somewhere, the Army wants troops to be able to run virtual scenarios on the same vehicles they'll actually fight with. All these characteristics are intertwined – and after its past troubles, the Army is acutely aware that maximizing one, such as protection, may compromise another, such as transportability. That's another thing the service wants feedback on, the announcement says: “The Army is interested in industry partners' ability to meet the desired characteristics and what trades” – that is, trade-offs – “may be necessary.” https://breakingdefense.com/2020/02/omfv-army-seeks-industry-advice-on-bradley-replacement

All news