The Air Force wants to think a little more creatively to help develop its plan to interconnect each of its sensors and weapons systems. It will look to industry, essentially crowd sourcing ideas to improve its Advanced Battle Management System (ABMS) architecture.
The Air Force recently put its existing model on SAM.gov with a Request for Information. It wants industry feedback by the end of December in hopes of improving the user experience through an industry-wide approach using systems engineering or system modeling language.
“We as warfighters are very focused on how does the user interact with all this information that’s made available to them? How do I transform all this data that is delivered, and then how do I turn it into knowledge?” Brig. Gen. Jeffery Valenzia, Air Force advance battle management system cross functional team lead, said in a recent webinar hosted by C4ISRNET.
ABMS is the Air Force element of the Joint All-Domain Command and Control initiative for joint force commanders to make decisions in real time with the multiple types of data available in a battle situation. With all five forces involved and potentially strategic partners as well, large amounts of data have to be turned into information that allows commanders to make lots of decisions quickly. Under the current system, data goes from sensors to a continental U.S.-based processing facility, and then the information gets sent back to commanders in the field. An interruption in the flow of information leaves commanders without the information they need to make decisions. An improved ABMS will distribute the data processing in new ways and make its delivery more secure.
”That current system, while very capable, is actually very difficult to scale and it’s very difficult if we get a communication line cut back to central processing,” said Valenzia.
The new ABMS digital infrastructure uses a cloud-based command and control application, and aerial edge network pods to deliver information to the five command and control nodes. This acquisition-approved model is what was put on SAM.gov for peer review and feedback. After discovering some problems with the original model, the Air Force wants to use the feedback from the RFI to fine-tune a mature, functional model for 2023.
“The reason why that’s significant is now industry has the exact description of what we’re modeling to and they have the objective measures of performance, so that they can understand what our current performance is. And they can look at how they can apply solutions to optimize that,” Valenzia said.
Getting the data analyzed and turned into useful information is only the first part of the equation. After information reaches battle managers, they have to use it to make decisions. The ABMS model needs a computer-generated assignment of priorities to rank the possible choices that battle commanders use to make decisions. An individual commander may know what he or she is supposed to do, but there may be a dozen different options in achieving that goal. ABMS software will use artificial intelligence to order decisions according to which have the most probability of success. Valenzia said this guidance would also level the field between commanders of varying operational ability.
“Computers do it really, really well. So we can start to look at what’s the error rate, we can look at variance that happens with a very low skill battle manager as opposed to a high skill battle manager within the same operational scenario,” he said.