NEWPORT, R.I. —
“Successful digital transformation requires development and integration of a wide range of capabilities to improve performance, unlock new opportunities, drive innovation, deliver new efficiencies, and inform analytic-based decision-making.”
—VADM William Galinis, Commander, Naval Sea Systems Command (NAVSEA)
A preview of NAVSEA Campaign Plan 3.0 reveals an emphasis on digital transformation. How this transformation will occur includes the employment of new processes that signal a cultural shift for the research and development (R&D) community. The new processes utilize Continuous Integration and Continuous Deployment (CI/CD) to streamline tasks and requirements in a cost-effective and expedient manner that will ultimately combine artificial intelligence (AI) and machine learning (ML) services with existing systems and the new services-based architecture applications under development today. One of the ultimate beneficiaries is the watchfloor operator for whom data will be automatically gathered, distilled, and presented as a user-friendly interface to allow for fast, informed, accurate, and certain decision-making. One area that will benefit greatly from a digital transformation is unmanned systems (UxS) and the command, control and communications (C3) systems that allow operators to interact with them.
Why is this important?
In a word: Scale.
A regional Command Task Force watchfloor has their command and control and Common Operation Picture to keep track of all Navy assets and missions in the area. Like air traffic control, operators are making sure the command structure is being followed. In this scenario, there is a hierarchy and known quantity of people and vessels.
When unmanned systems are added to the mix, the operational environment becomes significantly more complex. Future missions may have many vehicles and systems coming online at a rapid pace. If watchfloor manning is already a challenge, monitoring additional UxS traffic may become untenable under the current environment of inconsistent formats and incompatibility. The amount of data being collected from unmanned systems will likely overload the watchfloor and operators may struggle to maintain awareness.
Too much information is a liability. Understanding the struggles and challenges of watchfloor operators is essential for building the future Fleet. Digital transformation will help mitigate this problem.
At an Association for Unmanned Vehicle Systems International (AUVSI) conference in 2020, VADM Jim Kilby, Deputy Chief of Naval Operations for Warfighting Requirements and Capabilities (OPNAV N9), spoke about the technology needed to help operate UxS.
“Those enabling technologies are the naval tactical grid. So the ability to network and control, C2, all those unmanned vehicles is significant and important. Think about the aggregation of that demand on the network and understanding that and making sure we field a network that’s robust enough to handle all our vehicles in many different types of environments,” said Kilby.
“We view the world in the future of distributed maritime operations in a place where an air vehicle might have control of a surface vehicle and have to pass that control to a manned surface vessel or this unmanned operations center of the future,” said Kilby. “The ability to have a single control station that allows us to do that seamlessly will help us with training and the advancement and update of our systems in the future.”
Also at the AUVSI conference, Capt. Pete Small, program manager at the Unmanned Maritime Systems Program Office, said, “We’d like to have one interface that can support our family of unmanned systems across domains, whether that’s the aviation domain, the surface domain, the undersea domain. We do not want to be delivering this whole family of a variety of vehicles and platforms, each with its unique and different command and control interface, because that’s difficult – that’s a burden for the sailors to learn how to operate those things, it precludes interoperability between those systems and across domains, and it really increases the integration effort associated with our operations centers.”
That “one interface” is the Common Control System (CCS) and it is the operator software used to plan, execute, monitor, and analyze UxV missions that is being developed by Navy program offices and the Office of Naval Research (ONR) for their portfolio of unmanned systems. It enables the future Command, Control, and Communication (C3) for UxS. CCS not only allows for the rapid integration of new UxS capabilities, but its construct reduces cost and redundancies in software development. With the adoption of “DevSecOps,’ as new capabilities are added by individual efforts in the CCS community, the rest of the enterprise benefits as those capabilities can be shared and accredited for use.
The Concept: Development + Security + Operations = DevSecOps
DevSecOps is the concept of baking in cybersecurity practices into every step of the software development process. With DevSecOps, software engineers establish a “security as code” culture and collaborate on a common goal with cybersecurity teams. In order to eliminate the gaps that once existed between the information technology community and the cybersecurity community, DevSecOps is intended to quickly deliver cybersafe code. Stovepiped tasking is replaced with continuous inter-team communication and security requirements are instilled from the beginning through the end of software development and delivery.
The Foundation: Platform One
The Air Force established Platform One as a repository for sharing the tools and training to help others in the Department of Defense (DoD) set up “software factories’ and was designated as the DoD’s DevSecOps Enterprise Services Team. The idea that there is now a conveyor belt of software for building mission applications originated with the Air Force. Here, software consists of a collection of small reusable services, also known as micro services, that work together to form applications. Platform One allows software teams to employ DevSecOps and “helps instantiate DevSecOps CI/CD pipelines/software factories in days at various classification levels,” according to the Air Force’s Office of the Chief Software Officer.
Platform One is compliant with the DoD Enterprise DevSecOps Initiative (DSOP) and has DoD-wide reciprocity. It is part of a cultural shift in software development that provides teams with “collaboration tools, cybersecurity tools, source code repositories, artifact repositories, development tools, DevSecOps as a service, chats, etc.”
The Navy’s Digital Warfare Office (DWO) has stood up the Digital Integration Support Cell (DISC) to be the Department of the Navy service that provides the end-to-end solution, tools, and cloud resources, also known as Infrastructure as a Service (IaaS). It is meant to be the Navy-centric version of Platform One.
The Integration: RAIL
RAIL is the Rapid Autonomy Integration Lab, the Navy’s pilot program for autonomy development. The RAIL provides the digital engineering infrastructure, tools, and processes to rapidly develop, test, certify, and deploy new autonomous capabilities. RAIL will help scientists and engineers test and integrate new advances in autonomy software with existing unmanned vessels. This program not only lowers development costs but it speeds the integration of new systems into the Fleet. In this scenario, the DISC provides the conveyer belt and the RAIL is the software factory that is integrating the apps (services) that are traveling down the conveyer belt and being verified and validated in each stage of the pipeline.
If developers are using DevSecOps, Platform One, and RAIL, then they are doing the testing and getting the usability feedback from operators. This has the added benefit of providing true feedback without disrupting the operators. When paired with a strong simulation environment that is seamlessly integrated with the Live/Virtual/Constructive (LVC) environment, this enables the operator to run the equivalent of ‘gun-drills’ for UxS. Ultimately, LVC communities help supply this critical feedback.
In addition, the concept of DevSecOps combined with the tools from the DISC lets developers experiment freely to see what works and what doesn’t and also allows developers to fix the one component that’s slowing down the system rather than taking the entire system offline to fix a problem.
Most importantly, operator feedback is part of the continuous development process.
Keith Wichowski, engineer at the Naval Undersea Warfare Center (NUWC) Newport Division, has worked on C3 throughout his NUWC career during which he devised a program that provides C3 for UUVs and undersea distributed netted sensors for Navy program offices, ONR, the Defense Advanced Research Projects Agency (DARPA), and the submarine Fleet and is the one of the main programs used by the UUV squadron (UUVRON-1). Wichowski worked closely with UUVRON-1 to build the tools that they need for working with UUVs and distributed systems.
“My favorite part of the job has always been talking with the sailors and getting their feedback,” said Wichowski. “All of the engineers on my team stood watches with them in the UOC [Unmanned Operations Center] to get the sailors’ perspective of the tools they are developing. It was a very unique opportunity that I draw upon whenever I work with the next-generation efforts in the R&D community. When I worked on the Theater Anti-Submarine Warfare Offset Strategy to integrate and deploy a large-scale unmanned sensing system comprised of UUVs, USVs, and other systems in-theater, I did so with the understanding that the sailors would be responsible to stand watch for all of these disparate systems scattered across the sea. With all of these emerging technologies, the human-machine interface and operator workload has always been one of my main concerns while enabling these complex systems. From designing the software, choosing the hardware, implementing the cybersecurity controls, to striving to incorporate the subject matter expertise into the tools we build.”
Wichowski’s focus now is assisting with the Digital Transformation process. Wichowski has been focusing on digital infrastructure and human-autonomy teaming that leverages insights generated by Artificial Intelligence/Machine Learning from the vast data repositories (“data lakes”) that will be coming online.
“Because the number of operators cannot scale linearly with the number of distributed systems and/or number of data feeds, leveraging these intelligent agents supported by operator insight and oversight will be critical to enable the Navy of the future,” said Wichowski.
Culture shift for the R&D community
The UxS community can employ these processes to quickly provide the watchfloor operators with the tools and information they need to do their jobs efficiently and with greater accuracy. The culture shift comes as the R&D community adopts a new way of developing these systems.
The first step is understanding that the UxS community, and the Navy, is undergoing a transformation to allow for the expansion of capabilities. The R&D community continues its mission of integrating AI/ML into future systems to give operators the tools they need to do their jobs effectively without overwhelming them or overloading them. At the same time, the R&D community cannot forget the systems that already exist.
Incorporating established systems
It is important to note that as new systems are being developed, the goal is not to supplant established systems but rather incorporate them, help them evolve and carry out their mission. In order to be successful and expedient, there must be way forward for existing programs of record that are not slated to change.
According to Wichowski, one example is to leverage the work done by the LVC community to bridge the gap between established operational systems and bring those capabilities to the RAIL using common communication protocols.
Understanding what takes place on the watchfloor will inform all software and hardware development efforts. For example, at the beginning of any mission, operators consult data provided by the Commander Naval Meteorology and Oceanography Command (CNMOC). (In fact, CNMOC’s motto “It starts with us” refers to the fact that operators access their data at the start of their activity. It’s like turning on your GPS before you head out for a road trip.) The boats are relying on CNMOC’s data; the UxS community’s ability to tap into CNMOC’s vast data systems becomes crucial to new software development.
“In practice, failure of many or all of these variables is common, thereby reducing the general confidence by the system operator. Leadership and operators frequently ask for real-time observations to inform model-based automation, a fusion of relevant information for the present scenario to guide operator decisions. For example, when weather prediction models indicate that a powerful weather system is approaching the field of deployed gliders, the operator would like to see suggested route predictions, and the ability to see the proof — information sources — used by the algorithm,” said Wichowski.
“Systems that support UxS need to have common data structures, make use of LVC data, have both hardware and software in the loop, employ DevSecOps, AI/ML insights, common data lakes, and the RAIL,” said Wichowski. “Then, all of this data needs to be fed into distributed systems and incorporate logistics. Operators need to know what systems are fielded in the area. The success of these systems and the information they provide rely on a lot of things working together. Operators need to readily access information and that information must be human readable and machine usable.”
While continuing to support its existing user base, the current C3 products are working to extract the most successful components and convert them to services that can be leveraged by CCS and the Unmanned Maritime Autonomy Architecture community to enhance the warfighter capabilities and ultimately provide the warfighter with interfaces that are intuitive, informative, distributed, scalable, manageable (not overwhelming), and fast!
As a use case for employing DevSecOps, CCS, and RAIL, the UxS community is uniquely positioned for success because its systems are relatively new. As the culture shift and digital transformation take hold, other systems may follow. For developers, the cultural shift will include the concept of IaaS. Ideas conceived on a laptop can transition to a rack of servers and then quickly scale to be a part of a global system.
Ultimately, future software will incorporate multi-modal infographics and intelligent agents. Systems will be fast-acting, distributed, capable, and seamlessly integrate contextually relevant AI/ML insights for mission applications.
“The Navy needs the right infrastructure, intelligent agents, human-machine teaming, and C3 architecture to direct the coming fleets of autonomous systems, regardless of their mode of transportation. All of these elements and common data ontologies are necessary for it to work. Change is happening now and the community will need to work collaboratively to evolve quickly,” said Wichowski.
Knowing what works from a design sense and then knowing what is possible from a technical perspective is the key.
“Operator feedback is the critical piece,” said Wichowski. “In the end, there is always a human making the decision to push the button based on the information he or she was provided.”
NUWC Division Newport is a shore command of the U.S. Navy within the Naval Sea Systems Command, which engineers, builds and supports America’s fleet of ships and combat systems. NUWC Newport provides research, development, test and evaluation, engineering and fleet support for submarines, autonomous underwater systems, undersea offensive and defensive weapons systems, and countermeasures associated with undersea warfare.
NUWC Newport is the oldest warfare center in the country, tracing its heritage to the Naval Torpedo Station established on Goat Island in Newport Harbor in 1869. Commanded by Capt. Chad Hennings, NUWC Newport maintains major detachments in West Palm Beach, Florida, and Andros Island in the Bahamas, as well as test facilities at Seneca Lake and Fisher's Island, New York, Leesburg, Florida, and Dodge Pond, Connecticut.