Artificial Intelligence improves Soldiers' electronic warfare user interface
By Nancy Jones-Bonbrest, Army Rapid Capabilities and Critical Technologies Office
Aberdeen Proving Ground, Md. (March 18, 2019) -- The view for an Army electronic warfare (EW) Soldier can be daunting. Layers of data pour in from multiple sensors on the battlefield, while Soldiers work quickly to sort through the information, so they can determine what's enemy activity and what's simply "noise." At the same time, new data continues to arrive on the screen.
The complex mission executed by the Army's electronic warfare and Cyber Electromagnetic Activities (CEMA) Soldiers - which includes detecting and responding to enemy jamming attempts and other electronic interference - is intensifying. As the Army adds more sensors to its own equipment and the number and type of civilian signal-generating sources, including satellites and cell phones, continue to increase, an EW system's user interface can become crowded fast.
To reduce this cognitive burden and improve EW effectiveness for multi-domain operations, the Army Rapid Capabilities and Critical Technologies Office (RCCTO) is using artificial intelligence (AI) to quickly and accurately rank the incoming data in order of priority to the warfighter, who then has the option of filtering out the less important signals.
"With the current sensor technology, you end up with a lot of noises on your display," said SFC Steven Schoyen, an electronic warfare NCO for the 1st Stryker Brigade Combat Team (1st SBCT). "As the operator is attempting to discern what one correlation is, three or four more may begin to evolve. At the same time, noise and erroneous data is populating too."
The RCCTO is partnering with Soldiers from the 1st SBCT at Fort Wainwright, Alaska, who are using the new technology against operational scenarios. Their feedback will help improve effectiveness of the capability as the Army integrates it into EW systems.
The new expert learning AI prototype uses AI that is trained to reduce or eliminate common low-level tasks performed by EW Soldiers, while also simplifying the user interface of their battle management system.
"We know that many of the signals emanating from their display are less important signals, or signals that are known to be friendly or neutral sources," said Rob Monto, the director of the RCCTO Emerging Technologies Office. "Using AI, we can tell with extreme accuracy which signals are not important based on their pattern of life, their mobility profiles, and characteristics. By eliminating these signals, Soldiers can concentrate on those most significant."
The tool saves time by decluttering the user interface and enhancing Soldiers' ability to zero in on whether the emitter is from a "red" or enemy source, is a "blue" or friendly force signal, or just "gray" noise.
"Operating on the complex battlefields of the future or in megacities with dense radio frequency environments, as our EW Soldiers get more sensors, they'll also receive much more data," said Chief Warrant Officer 3 William Insch, an EW Technician with the Army's Project Manager Electronic Warfare & Cyber (PM EW&C). "AI will help translate all this data into information, and ultimately distill it into understanding, so EW Soldiers can better support the commander. And that's the key - we try to help the commander have better understanding of what the electromagnetic environment looks like from a friendly, neutral and adversarial perspective."
The Army RCCTO partnered with PM EW&C and the Combat Capabilities Development Command's C5ISR Center to develop the AI tool, which is planned for an operational pilot with a select forward deployed unit later this year.
Within the last year, the Army has delivered new electronic warfare prototype systems in response to an operational needs statement (ONS) from U.S. Army Europe. Soldiers with select units are using the equipment to implement electronic protection for their own formations, to detect and understand enemy activity in the electromagnetic spectrum, and to disrupt adversaries through electronic attack effects. These systems are interim solutions, designed as a bridge until the programs of record, including EW Planning Management Tool (EWPMT), can be fielded.
As part of the process, the RCCTO and PM EW&C worked with EW Soldiers to get feedback on the fielded prototypes and the soon-to-be-fielded EWPMT to drive design, performance, functionality and training. This approach of using direct user feedback enables the Army to be responsive to Soldier needs and rapidly insert new technology - like AI - as soon as it becomes available.
"The agile, iterative development of EWPMT is an outstanding way for EW and electromagnetic spectrum manager Soldiers to have a direct impact on the program," Insch said. "The engineers and developers are taking feedback from Soldiers, either through the ONS with the prototype equipment or with EW Soldiers visiting Aberdeen Proving Ground, and they immediately try to incorporate that feedback into the next capability drop."
The effort to prototype the AI tool began in October 2018 after the technology was discovered at an industry event. It was assessed by the EW community for scalability, ease of operations and security. In November the RCCTO, C5ISR Center and PM EW&C began a cognitive mapping process with the help of EW Soldiers like Insch and Schoyen, outlining the steps they take as they work through signal detection on the battlefield.
Now, the Army is running the technology through operational scenarios in labs prior to holding live demonstrations and evaluations this year in coordination with PM EW&C. Initial feedback demonstrates a significant reduction in the time it takes to generate the EW Soldiers' common operating picture. It also showed a reduction in the work these operators must perform, which decreases potential errors.
This is the second artificial intelligence and electronic warfare effort for the RCCTO, which recently conducted a Signal Classification Challenge to find an AI/Machine Learning application to help EW Soldiers on the back-end of signal detection. This new effort is on the front-end, applying AI to the user interface as signals are first displayed.