The world’s military forces have an urgent need for readily deployable real-time, real-world renderings and virtual databases supporting simulators which can be rapidly updated and protected from prying eyes and interference.
The ultimate aim is a ‘plug-and-play’ approach to virtual environments, merging real-world data sets and new tools from a variety of sources, distributed wherever they are needed, while ensuring the cybersecurity of all systems in the network.
Common content
To achieve that aim, the USAF announced the Simulator Common Architecture and Standards (SCARS) programme to establish a common open-systems architecture. It will be awarded to and managed by a prime contractor.
At the I/ITSEC 2019 event, held in December, Col Gerard Ryan, chief of the operational infrastructure division, Directorate of Training and Readiness, HQ USAF, said that the contractor-led standards initiative is essential to ensure swift adaptability of resilient systems.
He told media representatives at the time: ‘If we are to keep up with the pacing threats that are out there, we have to change our way of doing business. I’m not saying “take things away”. I’m saying we need to change this business model. And that theme is consistent throughout the air force. And I know the other services are saying that too.’
The idea is to follow the systems deployment practices of the commercial sector, while also ensuring higher security and performance reliability. ‘We certainly have done it in the cell phone world. We’ve done it in artificial intelligence. We’ve done it in so many other technological ways,’ Ryan added.
Steve Baigrie, CAE USA’s VP of operations and engineering, embraces the idea of standards-led plug-and-play design. CAE is prime contractor on a number of large USAF aircrew training services contracts (KC-135 tanker and C-130H transport) and has been working on simulator upgrades related to SCARS, particularly addressing cybersecurity.
CAE supports the SCARS approach to training systems development. ‘Standards are important because the government wants plug-and-play functionality,’ Baigrie explained, ‘so that if a new visualisation tool comes up, you can use their data, or if a new widget comes along, you can use the tool. This is all the leading edge of synthetic environments and it’s the trend that DoD is going on.
‘SCARS is a vision the US Air Force has for a common architecture, a plug-and-play architecture, similar to what we see in the commercial industry, where you can just download an app on your phone and use it.’
Working group
While the air force’s selection of system integrator for SCARS is still pending, an industry working group, in which CAE participates, has been actively addressing some of the foundational needs of the future plug-and-play infrastructure for the past year and a half and has presented concepts to the USAF for cyber and common architecture.
‘The first thing that they want to tackle is cyber,’ Baigrie told Shephard. ‘Common architecture is more of a future effort, but cyber is a more pressing problem right now. There are many legacy simulators out there that need cybersecurity or need to improve their cyber posture. We’ve been working on cyber implementation of cybersecurity on several platforms that we are part of.’
CAE is prime contractor on the KC-135 Aircrew Training System (ATS) and also supports Lockheed Martin as a subcontractor for the C-130J ATS, providing cybersecurity support among other things. The company supplies a solution called CCMS (Centralized Cybersecurity Management System) which offers various tools and capabilities, including the ability to push and manage changes or patches.
‘It has a lot of capabilities like authentication of parties when you log in, vulnerability assessment, and it implements some DoD-mandated cyber tools and host-based security systems – certain things like compliance checks and intrusion prevention, data loss prevention, rogue system prevention,’ Baigrie said.
CAE developed CCMS for USN programmes and has adapted it to air force environments; it has already been deployed on the KC-135 and C-130J ATS.
‘It’s actually working and proven,’ Baigrie said. ‘The whole idea is to improve the cyber posture of their entire training system and to allow them to manage it from a non-technical user point of view. We’re going to be implementing that on a C-130H, for a programme that we won in September 2018. They’re going to be contracting us for cyber updates under SCARS.’
Cyber challenge
The challenge is keeping different systems, at various locations, in line with the same stringent functionality and security standards. With the KC-135 ATS, for example, there are 19 full-mission aircraft system simulators at 13 sites and other training devices attached to the programme – such as boom operator trainers – all of which need to be cyber secure across a distributed network.
The challenge is keeping different systems, at various locations, in line with the same stringent functionality and security standards.
‘The old-fashioned brute-force way is to have someone on site that is able to patch things and adapt to the existing DoD requirements for cyber,’ Baigrie said.
‘That’s very labour-intensive and it’s problematic. If they don’t do the update right, the simulator could be non-functional. It’s a significant problem and you have to remember that the devices are all networked because they do what they call distributed mission operations training. That’s multiple devices networked together from different locations and even different types of devices and different platforms – like C-17 and KC-135 – practising a mission. That’s all networked.’
The CCMS facilitates management of the whole enterprise from a common, single, remote-access point for scanning and centralised management of all cybersecurity services. This is usually a TSSC (training services support centre) with technicians and engineers available to keep the enterprise running properly, Baigrie said.
‘They are not cyber experts, but they can operate the system,’ he told Shephard. ‘It gives them a lot higher cyber posture and a better way to manage their training enterprise.’
Cybersecurity is essential given the training rhythm required for crew proficiency, with simulators active for up to 16h a day, but it is also critical to ensure that vital intelligence contained in the simulators remains secure.
‘They don’t want their tactics and techniques and operating procedures disclosed outside of the DoD, or even outside the air force,’ Baigrie explained. An intruder could potentially also piece together mission-sensitive information – including plans and locations – based on the data loaded into the simulators. And cybersecurity must be ensured while networking a suite of systems that may include legacy technologies deployed over the past two decades.
Flexible realism
The other priority for both SCARS and the US Army’s Synthetic Training Environment (STE) programme is flexibility and realism – the ability to load real-world scenarios quickly and reliably, which can be used both in training and in strategic applications. CAE has also been developing tools to support rapid deployment of synthetic environments.
‘There are a lot of requirements where, given the changing situation in the world, they want to have these synthetic environments updated quickly,’ Baigrie said. ‘The historic way of doing things is a very labour-intensive process where you get imagery and you process it, and you create a database based on it, but that can be a year-long or 18-month-long activity.’
That timeframe no longer works under modern operating conditions. ‘We have tools and capabilities for near-real-time updates,’ Baigrie said. CAE’s systems support what he describes as ‘data fusion’, taking data from sensors updated in real time, combined with data from synthetic environments, to create a rapidly deployed representation of real-world conditions.
‘Distributing it to the device that they want to see it on is another challenge,’ Baigrie said. CAE is developing systems, tools and processes through which users can adjust their synthetic environment to operational needs and also visualise it whenever and wherever they want.
‘That requires a lot,’ Baigrie told Shephard. ‘You’re talking about automated workflows and analytics, machine learning, machine-to-machine interfaces, artificial intelligence and data management –we’re talking about massive amounts of data. For distribution, you are talking about cloud-enabled architecture that allows you to distribute it to the point of need. Providing that capability is a giant leap forward from where traditionally synthetic environments have been in the past.’
CAE has contracts in place with the National Geospatial-Intelligence Agency and the US Special Operations Command for rapid generation of synthetic environments, developed through the company’s Common Database (CDB) architecture. This functions as a geospatial data repository which can supply relevant information for modelling, through the cloud, within minutes or hours, rather than weeks or months. Data models can be delivered to simulators or commercial laptop computers, to support both training and mission preview applications.
Geospatial standard
The independent, international standards development organisation Open Geospatial Consortium (OGC) approved CDB as a standard for geospatial data in September 2016, naming it OGC CDB. It has been adopted by other militaries for training and simulation applications, including the Canadian Coast Guard, the German Army’s aviation school at Bückeburg, the UK’s Military Flying Training System and the Turkish Air Force.
‘The actual management of the data is one of the big research areas that we’re working on – how the data is exchanged, how it is stored, developing a data hub, so that it can be sent from a central location,’ Baigrie said. ‘You’re talking about terabytes of data – a high-resolution area can be terabytes – and you multiply that around the world. It becomes very important how you manage that data, what the latest configuration is and how you keep it up to date.’
This approach, which Baigrie describes as ‘data fusion’, enhances training capabilities and can also support strategic planning.
‘Outside the US, the common synthetic environment is something that many governments are looking at as a tool for training, mission rehearsal, decision support, course of action analysis,’ he said. ‘We are involved with the UK on a project that they call a single synthetic environment, doing some technology development and assessment to see how a single synthetic environment can support the whole government, including decision support tools, course of action analysis, helping to make decisions at a government level faster in real-time…
‘You can provide a common operating picture of the scenario that is unfolding, including elements in the field that are part of that scenario, running faster than real-time analysis of potential outcomes, so you can determine a course of action and recommendation. That’s requiring a lot of advanced analytics and artificial intelligence. That’s something we’re working on, not just in the US but globally, and at different locations around the world.’
Limited benefits?
While standards are beneficial to rapid deployment and adaptability of systems, not everyone agrees that SCARS alone will deliver the desired outcomes. The horse may be out of the barn, but W Garth Smith, president of MetaVR, believes there is still room to adapt the saddle.
Not everyone agrees that SCARS alone will deliver the desired outcomes.
‘There is significant value in the ultimate goal of the SCARS programme – to develop common architecture requirements and standards for US Air Force simulators – but I’m not convinced that the process by which they hope to arrive at these standards is the right approach,’ he told Shephard.
‘Historically, the successful development of simulator interoperability standards has been led by specialists in the field, rather than via the “top-down” approach being taken with SCARS. For example, if you look at the 1980s DARPA programme called SIMNET, which resulted in the DIS protocols, the government picked a small company to design and implement a networked virtual multi-player environment, and the resulting SIMNET network protocols were pushed out to the community as a public standard for conducting real-time platform-level wargaming across multiple host computers used worldwide.
‘Under SCARS, the government is doing the opposite: seeking standards to be developed by a DoD prime, and my view is that this won’t result in the best outcome for industry or the end-user. There are existing IG [image generator] COTS solutions with innovative terrain formats on the market right now, developed by specialist companies, that could be put through a competitive downselection process to find the best fit for their needs. The best solution could then be mandated as a standard by giving the IG source code to any DoD company that is to deliver a simulator. This would ensure that the organisations with the right skills are leading the development of IG content, which is ultimately the gauge by which simulator usability and training value are measured by the end-customer.’
Once the SCARS contract has been awarded, it will be interesting to see if the model chosen for its implementation delivers the results the air force requires.