16 octobre 2018 | International, Naval
By: Daniel Cebul
WASHINGTON — The Office of Naval Research awarded Lockheed Martin Oct. 1 a two-year, $5.8 million contract to explore how machine learning and artificial intelligence can make complex 3-D printing more reliable and save hours of tedious post-production inspections.
In today's factories, 3-D printing parts requires persistent monitoring by specialists to ensure intricate parts are produced without impurities and imperfections that can compromise the integrity of the part overall. To improve this laborious process, the Navy is tasking Lockheed Martin with developing multi-axis robots that use lasers to deposit material and oversee the printing of parts.
Lockheed Martin has multiple partners on the contract including Carnegie Mellon University, Iowa State University, Colorado School of Mines, America Makes, GKN and Wolf Robotics and Oak Ridge National Laboratory.
The contract covers what Glynn Adams, a senior engineer with Lockheed Martin, describes as the pre-flight model of the program's development. Initial work will focus on developing computer models that can predict the microstructures and mechanical properties of 3-D printed materials to generate simulation data to train with. Adams said the Carnegie Mellon team will look at variables such as, “the spot size of the laser beam, the rate of feed of the titanium wire [and]the total amount energy density input into the material while it is being manufactured.” This information helps the team predict the microstructure, or organizational structure of a material on a very small scale, that influences the physical properties of the additive manufactured part.
This data will then be shared with Iowa State, who will plug the information into a model that predicts the mechanical properties of the printed component. By taking temperature and spot size measurements, the team can also ensure they are, “accurately controlling energy density, the power of both the laser and the hot wire that goes into the process,” Adams said..
“All of that is happening before you actually try to do any kind of machine learning or artificial neural networks with the robot itself. That's just to try to train the models to the point where we have confidence in the models,” Adams said.
Sounds easy, right?
But one key problem could come in cleaning up the data and removing excess noise from the measurements.
“Thermal measurements are pretty easy and not data intensive, but when you start looking at optical measurements you can collect just an enormous amount of data that is difficult to manage,” Adams explained. Lockheed Martin wants to learn how shrink the size of that dataset without sacrificing key parameters. The Colorado School of Mines and America Makes will tackle the problem of compressing and manipulating this data to extract the key information needed to train the algorithms.
After this work has been completed, the algorithms then will be sent to Oak Ridge National Laboratory, where robots will begin producing 3-D titanium parts and learn how to reliably construct geometrically and structurally sound parts. This portion of the program will confront challenges from the additive manufacturing and AI components of the project.
On the additive manufacturing side, the team will work with new manufacturing process, “trying to understand exactly what the primary, secondary and tertiary interactions are between all those different process parameters,” Adams said. “If you think about it, as you are building the part depending on the geometric complexity, now those interactions change based on the path the robot has to take to manufacture that part. One of the biggest challenges is going to be to understand exactly which of those parameters are the primary, which are the tertiary and to what level of control we need to be able to manipulate or control those process parameters in order to generate the confidence in the parts that we want.”
At the same time, researchers also will tackle AI machine learning challenges. Like with other AI programs, it's crucial the algorithm is learning the right information, the right way. The models will give the algorithms a good starting point, but Adams said this will be an iterative process that depends on the algorithm's ability to self-correct. “At some point, there are some inaccuracies that could come into that model,” Adams explained. “So now, the system itself has to understand it may be getting into a regime that is not going to produce the mechanical properties or microstructures that you want, and be able to self-correct to make certain that instead of going into that regime it goes into a regime that produces the geometric part that you want.”
With a complete algorithm that can be trusted to produce structurally sound 3-D printed parts, time-consuming post-production inspections will become a thing of the past. Instead of nondestructive inspections and evaluations, if you “have enough control on the process, enough in situ measurements, enough models to show that that process and the robot performed exactly as you thought it would, and produced a part that you know what its capabilities are going to be, you can immediately deploy that part,” said Adams. “That's the end game, that's what we're trying to get to, is to build the quality into the part instead of inspecting it in afterwards."
Confidence in 3-D printed parts could have dramatic consequences for soldiers are across the services. As opposed to waiting for replacement parts, service members could readily search a database of components, find the part they need and have a replacement they can trust in hours rather than days or weeks. “When you can trust a robotic system to make a quality part, that opens the door to who can build usable parts and where you build them,” said Zach Loftus, Lockheed Martin Fellow for additive manufacturing. “Think about sustainment and how a maintainer can print a replacement part at sea, or a mechanic print a replacement part for a truck deep in the desert. This takes 3-D printing to the next, big step of deployment.”
25 novembre 2024 | International, Aérospatial
The US Department of Defense (DoD) is considering revising the 2022 Nuclear Posture Review to maintain effective nuclear deterrence amid perceived growing threats from China and Russia and the potential absence of nuclear arms control agreements after February 2025.
22 juillet 2020 | International, C4ISR
Mark Pomerleau One of the clearest examples of how the military wants to defeat adversaries using information warfare is by publicly disclosing what those enemies have been doing and what capabilities they have. Information warfare can be abstract, combining cyber, intelligence, electronic warfare, information operations, psychological operations or military deception as a way to influence the information environment or change the way an adversary think. “At our level, the most important thing we can do is to be able to expose what an adversary is doing that we consider to be malign activity, in a way that allows that to be put in the information environment so that now more scrutiny can be applied to it,” Lt. Gen. Timothy Haugh, commander 16th Air Force, the Air Force's newly established information warfare organization, told reporters during a media round table in late February. One of the first ways the Department of Defense has sought to test this is through U.S. Cyber Command's posting of malware samples to the public resource VirusTotal. Malware samples discovered in the course of operations by the Cyber National Mission Force are posted to the site to inform network owners. It also helps antivirus organizations of the strains build patches against that code and helps identify the enemies' tools being used in ongoing campaigns. Haugh, who most recently led the Cyber National Mission Force, explained how these cyber teams, conducting what Cyber Command calls hunt forward operations, were able to expose Russian tactics. U.S. military teams deploy to other nations to help them defend against malign cyber activity inside their networks. “Those defensive teams then were able to identify tools that were on networks and publicly disclose them, [and] industry later attributed to being Russian tools,” he said. “That was a means for us to use our unique authorities outside the United States to be able to then identify adversary activity and publicly disclose it.” Officials have said this approach changes the calculus of adversaries while also taking their tools off the battlefield. “Disclosure is more than just revealing adversary intent and capabilities. From a cyberspace perspective, disclosure is cost imposing as it removes adversary weapons from the ‘battlefield' and forces them to expend resources to create new weapons,” Col. Brian Russell, the commander of II Marine Expeditionary Force Information Group, told C4ISRNET in June. “Disclosure forces the adversary to ask: ‘How were those capabilities discovered?' It causes them to investigate the cause of the disclosure, forcing them to spend time on something other than attacking us. If I can plant a seed of doubt (messaging) that the disclosure might have been caused by someone working on the inside, it makes them question the system's very nature, perhaps spending more time and resources to fix the system.” The NSA has demonstrated a similar tactic when it created its cybersecurity directorate in late 2019. The entity was formed in part, due to the fact that adversaries were using cyberspace to achieve strategic objectives below the threshold of armed conflict. Now, the directorate uses its intelligence and cyber expertise to issue advisories to the network owners of cybersecurity threats so they can take the necessary steps to defend themselves. One recent advisory had direct bearing on a nation state's malicious activity, according to a senior intelligence official. In late May, the agency issued an advisory regarding a vulnerability in Exim mail transfer agent, which was being widely exploited by a potent entity of Russia's military intelligence arm the GRU called Sandworm. “Quickly thereafter, we saw five cybersecurity companies jumped on it and really used that to deepen and expand and publish information about the GRU's infrastructure that they use to conduct their cyberattacks and further information as well,” the official told reporters in early July. “That was terrific because we felt that that had a direct impact on a major nation state in terms of exposing their infrastructure ... and we saw significant patch rates go up on a vulnerability that we knew they were using. That's the kind of thing that we're looking for.” The military has had to think differently to combat for how adversaries are operating. “A central challenge today is that our adversaries compete below the threshold of armed conflict, without triggering the hostilities for which DoD has traditionally prepared,” Gen. Paul Nakasone, commander of Cyber Command, wrote in prepared testimony before the House Armed Services Committee in early March. “That short-of-war competition features cyber and information operations employed by nations in ways that bypass America's conventional military strengths.” These disclosures or efforts to call out malign behavior have also taken the forms of media interviews and press releases. For example, Gen. Jay Raymond, the head of U.S. Space Command and the commandant of Space Force, said in a February interview in which he detailed what he deemed unacceptable behavior by Russia in space, a surprising charge given how tight lipped the U.S. government typically is about its satellites. “We view this behavior as unusual and disturbing,” he said of Russian satellites creeping up to American ones. “It has the potential to create a dangerous situation in space.” Or consider that leaders from Africa Command on July 15 issued a press release detailing the activities of the Wagner Group, a Russian security company, as acting on behalf of the Russian state to undermine the security situation in Libya. “U.S. Africa Command (AFRICOM) has clear evidence that Russian employed, state-sponsored Wagner Group laid landmines and improvised explosive devices (IEDs) in and around Tripoli, further violating the United Nations arms embargo and endangering the lives of innocent Libyans,” the release said. “Verified photographic evidence shows indiscriminately placed booby-traps and minefields around the outskirts of Tripoli down to Sirte since mid-June. These weapons are assessed to have been introduced into Libya by the Wagner Group.” Moreover, Africa Command's director of operations called out Russia, noting that country's leaders have the power to stop the Wagner Group, but not the will. Sixteenth Air Force, at the request of C4ISRNET, provided a vignette of such behavior from Russia in the form of how it covered up the explosion of a radioactive rocket, dubbed Skyfall. According to the service, Russia took extreme steps to curb monitoring of the site where the explosion took place and sought to conceal the true nature of the explosion potentially hindering surrounding civilian populations from receiving adequate medical treatment and guidance. With new forces integrated under a single commander, using unique authorities to collect intelligence and authorities to disclose, 16th Air Force is now better postured to expose this type of malign activity, which previously the U.S. government just didn't do. Top Pentagon leaders have explained that the dynamic information warfare space requires a new way of thinking. “We've got to think differently. We've got to be proactive and not reactive with messaging,” Lt. Gen. Lori Reynolds, the Marine Corps' deputy commandant for information, told C4ISRNET in an interview in March. “We have been very risk averse with regard to the information that we have. You can't deter anybody if you're the only one who knows that you have a capability.” https://www.c4isrnet.com/information-warfare/2020/07/20/the-new-ways-the-military-is-fighting-against-information-warfare-tactics/
4 novembre 2024 | International, C4ISR, Sécurité