January 4, 2024 | International, C4ISR
August 6, 2019 | International, C4ISR
Can a dragonfly teach a missile how to hunt?
By: Jen Judson
WASHINGTON — A computational neuroscientist is studying whether a dragonfly's excellent hunting skills can be replicated in a missile's ability to maneuver and destroy targets midair with better precision.
Dragonflies are vicious little creatures with a hit-to-kill track record of 95 percent, meaning only 5 percent of its prey escapes.
Sandia National Laboratories' Frances Chance is building algorithms that simulate how a dragonfly processes information when intercepting prey, and she's testing them in a virtual environment. So far, the results are promising.
The laboratories are federally funded and focus on national security missions through scientific and engineering research. The project is a yearlong, high-risk, high-gain effort that will wrap up in September, and it is funded by Sandia's Autonomy for Hypersonics Mission Campaign, Chance said.
“I think what is really interesting about insects, in general, is they do something really fast and really well, but they are not particularly smart in the way you or I would think of ourselves as being smart,” Chance told Defense News in a recent interview.
While insects may not be the right fit for studying cognitive capabilities to develop complex artificial intelligence, they are ideal for developing efficient computations for intercept capability. A dragonfly can react to a particular prey's maneuvers in 50 milliseconds, Chance explained. That amount of time accounts for information to cross three neurons in a dragonfly's brain. This indicates the dragonfly doesn't learn how to hunt, but rather the skill is inherent and part of its brain's hard-wiring.
“The challenge then is: Is there anything that we can learn from how dragonflies do this that we can then bring to the next generation of missiles, or maybe even the next-next generation of missiles?” Chance said.
By developing an artificial neural network that mimics a dragonfly's ability to hunt and then applying it to missile capabilities that rely on computation-heavy systems, one could reduce the size, weight and power needed for a missile's onboard computers; improve intercept techniques for targets such as hypersonic weapons; and home in on targets using simpler sensors.
If the model of a dragonfly's neural circuit developed through Chance's research shows enough promise, she would then pass the information to scientists, who would try to directly apply it to weapons systems.
One of the greatest leaps involves adapting an algorithm to handle the speed at which a missile flies. While a dragonfly is fast, it's not nearly as fast as a missile. Animal brains process information significantly slower than a computer, so it's possible computations can be sped up to better align with the speed at which a missile approaches targets.
“The hope is that even if the algorithm isn't wildly successful, you might be able to say something about what you can get away with in terms of what types of capabilities you give the next generation of weapons,” Chance said.
The model she's building is several steps removed from implementation onto a weapon. “I would consider the project complete when we have a viable model — ‘viable' meaning it does interception — and a bonus if it's neurobiologically plausible. There is no reason to force that for this type of research, but only because it doesn't necessarily matter; so something biologically inspired that works I would consider a success.”
https://www.c4isrnet.com/land/2019/08/05/can-a-dragonfly-teach-a-missile-how-to-hunt/
On the same subject
-
-
January 27, 2020 | International, Land
Army picks 6 to work on autoloader for extended-range cannon
By: Jen Judson WASHINGTON — The Army has picked six companies to work on concepts and designs for an autoloader for the service's future Extended-Range Cannon Artillery (ERCA) program currently under development, according to a Jan. 24 Army Futures Command statement. While the first ERCA cannons will be fielded in fiscal 2023, the goal is to begin fielding the system with an autoloader just one year later. The companies — Actuate (formerly Aegis Systems, Inc.); Apptronik, Inc.; Carnegie Robotics LLC; Pratt & Miller Engineering; Neya Systems, LLC and Hivemapper, Inc. — will work under the Army Capability Accelerator and the Army Applications Laboratory (AAL) as part of the Field Artillery Autonomous Resupply (FAAR) “cohort” and will come up with novel, outside-of-the-box concepts for the autoloader. AAL is part of AFC, the Army's new four-star command in charge of rapid modernization that will align with the service's new developing doctrine. The cohort began work on Jan. 13 in Austin, Texas, where the AAL and AFC reside, and will wrap up work with capability presentations on April 2, the statement notes. “Sourced from across the country, the selected companies represent a range of technologies and expertise all aimed at developing autonomous resupply capabilities,” the statement reads. Among the companies selected, Actuate specializes in artificial intelligence focusing on computer vision software that turns any security camera into an “intruder- and threat-detecting smart camera," the release states. Apptronik is a robotics company spun out of the Human Centered Robotics Lab at the University of Texas at Austin. Pittsburgh-based Carnegie Robotics specializes in robotic sensors and platforms for defense, agriculture, mining, infrastructure and energy applications and was founded out of Carnegie Mellon University's National Robotics Engineering Center. Pratt & Miller's focus has been on addressing technology challenges in the motorsports, defense and mobility industries. Neya Systems, also from Pittsburgh, Pennsylvania, is another robotics company focused on advanced unmanned systems, off-road autonomy and self-driving vehicle technologies. The AAL has become the face of doing business with the Army in the startup community and has set up shop in the heart of Austin within an innovation incubator hub called the Capital Factory. Anyone can walk through an open garage door and pitch ideas to the Army and the service. But the Army is also going out to companies and trying to convey problems they need solved on the battlefield in the hopes of finding new and novel solutions. “Designed for small businesses and companies that don't typically work with the federal government, the program connects qualified companies that want to grow a new line of business into the DoD with Army stakeholders who want to speed capability development, transition to a program of record, or de-risk and inform requirements,” according to the statement. “We've spent the past year working to introduce commercial business models that translate to the Army and can help evolve its approach to capability development,” Porter Orr, product innovation lead at AAL, said. “We're helping nontraditional companies build a new line of business into the government. And that's important, but it's just as important that we're giving Army leaders a choice between writing a large check or doing nothing. This is a way for them to get more insight—more confidence—in a solution before purchasing it. That will mean a higher probability of success in the field.” Cohort participants receive $150,000 to complete a 12-week program ending in a pitch to the Army. FAAR is the pilot effort of likely many attempts to bring in non-traditional businesses to help solve some of the Army's problems both big and small. https://www.defensenews.com/land/2020/01/24/army-picks-6-to-work-on-autoloader-for-extended-range-cannon
-
July 31, 2020 | International, Aerospace
U.S. Air Force Launches Three-Year Fielding Plan For Skyborg Weapons
Steve Trimble July 07, 2020 The next combat aircraft to enter the U.S. Air Force inventory will not be a manned sixth-generation fighter or even the Northrop Grumman B-21. By fiscal 2023, the Air Force expects to deliver the first operational versions of a new unmanned aircraft system (UAS) called Skyborg, a provocative portmanteau blending the medium of flight with the contraction for a cybernetic organism. The Skyborg family of aircraft is expected to fill an emerging “attritable” category for combat aircraft that blurs the line between a reusable UAS and a single-use cruise missile. July 8 award date for Skyborg contracts Leidos is managing autonomy mission system As the aircraft are developed, Skyborg also will serve as the test case of a radical change in acquisition philosophy, with ecosystems of collaborative software coders and aircraft manufacturers replacing the traditional approach with a supply chain defined by a single prime contractor. The Air Force also plans to manage the Skyborg aircraft differently than other UAS. Although Air Combat Command (ACC) is considering the Skyborg family as a replacement for pre-Block F-16s after 2025 and MQ-9s after 2030, the aircraft is not likely to fit neatly into an existing force structure with dedicated Skyborg squadrons. “Even though we call Skyborg an attritable aircraft, I think we'll think of them more like reusable weapons,” says Will Roper, assistant secretary of the Air Force for acquisition, technology and logistics. The Skyborg is an attritable weapon, which means key components such as the jet engine will be designed with a short service life. Credit: AFRL via YouTube The Skyborg propulsion systems—including expendable subsonic and supersonic jet engines—will be rated with a fraction of the service life expected of a fully reusable UAS or manned aircraft. “We'll do whatever number of takeoffs and landings they're ‘spec'd' for, and then we'll attrit them out of the force as targets and just buy them at a steady rate,” Roper says. Starting in fiscal 2023, a concept of operations for a formation of four Lockheed Martin F-22s will include Skyborgs as part of the manned aircraft's load-out. “I expect that the pilots, depending on the mission, [will] decide: Does the Skyborg return and land with them and then go to fight another day, or is it the end of its life and it's going to go on a one-way mission?” Roper explains. In some cases, the pilot may decide a target is important enough that it is worth the loss of a Skyborg, even if its service life has not been used up, he adds. As the concept evolves, a diverse array of Skyborg aircraft designs will likely find roles beyond the air combat community, Roper says. “I don't think it'll just be fighters,” he says. “I think they'll fly with bombers. I think they'll fly with tankers to provide extra defensive capability. That's what I love about their versatility and the fact that we can take risks with them.” Skyborg is often presented as the epitome of the “loyal wingman” concept, in which one or multiple UAS are controlled or managed by a manned aircraft to perform a variety of surveillance, support and strike tasks during a mission. But the aircraft also could have the ability to operate independently of a manned aircraft, with the capability to launch and recover hundreds of such systems without the need for runways or even bases. The Kratos XQ-58A, which achieved first flight in March 2019, is one of several potential members of the Skyborg UAS family. Credit: U.S. Air Force “If [China and Russia] know that they have to target only tens or even hundreds of ports and airfields, we have simplified their problem,” says ACC chief Gen. Mike Holmes. The new class of attritable aircraft, he says, are designed so that “we can still provide relevant high-tempo combat power to be freed up from a runway.” If Skyborg is the future, it begins on July 8. The Air Force Research Laboratory (AFRL) is due on the second Wednesday of this month to award a contract to start developing the first in a family of experimental UAS bearing the name Skyborg. The AFRL already has a stable of potential concepts. The Kratos XQ-58A Valkyrie, which has flown four times since March 2019, is the most visible example of the AFRL's Low-Cost Attritable Aircraft Technology platform. Meanwhile, the Low-Cost Attritable Aircraft Platform--Sharing project quietly kept several UAS industry leaders involved in design studies, including Boeing, General Atomics Aeronautical Systems Inc., Lockheed Martin and Northrop Grumman. Each company selected will be awarded a contract with a maximum value of $400 million over a five-year ordering period. But the core of the Skyborg program is the software; specifically, the military aviation equivalent of the algorithm-fed convolutional neural networks that help driverless cars navigate on city streets. In announcing Leidos on May 18 as the Skyborg Design Agent (SDA), the AFRL selected the same company that delivered the software “brain” of the Navy's Sea Hunter unmanned surface vehicle, which navigated from San Diego to Honolulu in 2018. As SDA, Leidos' role is to deliver a software core that uses artificial intelligence to learn and adapt as the aircraft flies. The autonomy mission system core—as integrated by Leidos from a combination of industry and government sources—will be inserted into multiple low-cost UAS designed by different companies, with each configured to perform a different mission or set of missions. That is how the Skyborg program is set up today, but that is not how it started. Roper created the original “Skyborg” term and concept when he led the Strategic Capabilities Office within the Office of the Secretary of Defense in 2012-17. Roper transferred Skyborg to the AFRL, where it was renamed Avatar. A year after taking over Air Force acquisition in 2017, Roper changed the name back to Skyborg and created a program office in October 2018. In March 2019, Roper revealed the Skyborg concept to a group of reporters a week before the AFRL issued the first request for information to industry about the program. At that time, Skyborg was still organized more traditionally, with plans to select a single contractor to serve as a prime integrator. By early 2020, program officials reorganized Skyborg into modular hardware and software subcomponents built on an open architecture that requires no prime integrator. As the acquisition strategy has evolved, so has the Air Force's thinking about how to use the Skyborg family of systems. “The whole idea was [that] the contested environment is going to be challenging, it's going to be uncertain, and so it makes the most sense to have something that doesn't have a pilot in it to go into the battlefield first,” Roper says. “But once you agree that's a self-evident operational concept, it opens up the door for a lot of nontraditional thinking for the Air Force.” After a 2-3 year experimental phase, the AFRL plans to deliver an early operational capability in fiscal 2023. Follow-on operational Skyborgs could be funded within the Next-Generation Air Dominance (NGAD) project or through a separate program of record. The Skyborg concept even has links to the Air Force's architecture for the Advanced Battle-Management System (ABMS). “Attritable-ONE,” which is defined as having “multirole attritable capabilities,” is one of about 30 product lines in the ABMS architecture. “Skyborg and the AttritableONE teams are closely coordinated for planning and collaboration purposes,” the AFRL informed industry in response to questions about the Skyborg solicitation. The aircraft supplier must deliver a highly flexible design. Leidos, the design agent, will provide the autonomous mission system that will serve as the pilot, flight control computer and mission systems operator for the aircraft. But the “size, weight, power and cooling details for the Skyborg core autonomy system have not been finalized,” the AFRL told the bidding companies. “The majority of the system will be software-based and integrate with the sensors onboard the host aircraft,” the AFRL says. “Extensive collaboration between the Skyborg system design agent and the participants in this [contract] is expected.” https://aviationweek.com/ad-week/us-air-force-launches-three-year-fielding-plan-skyborg-weapons