Knowing what’s going to happen before it does has long been an aspiration; it’s been the subject of blockbuster movies, books and TV series. But in reality, its potential has never been unlocked – until now. Modelling and simulation, data analytics, AI and machine learning, distributed systems, social dynamics and human behaviour simulation are fast becoming the go-to tools, and their advantages could offer significant advantages for the battlespace of tomorrow.
Since 2018, London-based technology provider Improbable has been working closely with the UK Ministry of Defence (MoD) to explore the utility of synthetic environments (SEs) for tactical training and operational and strategic planning.
The company has a number of contracts to provide SEs to the MoD for a range of uses. At the core of this work is Skyral, a platform that supports an ecosystem of industry and academia enabling the fast construction of new SEs for almost any scenario, using digital entities, algorithms, AI, historic and real-time data.
Experts in modelling and simulation from the universities of Oxford, Bristol, Leeds, Manchester and Durham have teamed up with Improbable as part of a new initiative – the Myridian programme – to help the UK analyse and overcome what it calls “fluid, fast-moving and intricately interrelated challenges” from climate change and humanitarian crises to political interference, disinformation and threats posed by potential adversaries across air, land, sea, space and cyber.
Wespoke with Rob Solly, director of research partnerships for defence at Improbable, to find out what the project’s aims are and how it could help government and other public bodies assess threats and respond in a timely and effective manner. Solly was previously the head of the Defence and Security Analysis Division at the UK MoD’s Defence, Science and Technology Laboratory (Dstl) before joining Improbable in March 2021.
Andrew Tunnicliffe (AT): Tell us a little about the initiative – what it is?
Rob Solly (RS): Myridian is a programme to deliver closer alignment and collaboration between and across government, industry and the research community to build an SE ecosystem that acts as a strategic resource for the UK in the defence, national security and resilience sectors. An SE ecosystem will bring together the best qualities of government, industry and research, driving innovation, capability transformation and therefore strategic advantage.
AT: Who is involved and what is the thinking behind it?
RS: Improbable has set up the Myridian programme with an initial group of partners: the Alan Turing Institute; and Oxford, Bristol, Leeds, Manchester, Durham and Coventry Universities. Each of these brings world-class expertise in topics related to SEs, including developing models of some of the most difficult aspects of human behaviour (human geography, behavioural science, economic activity and societal resilience), harnessing AI to bring these models to life, mathematical modelling and computer science expertise to help integrate these models, and statistical techniques to calibrate them and test their validity.
AT: Does the MoD have an active role?
RS: The MoD is probably the largest user of SEs of all government departments, with applications ranging from the familiar (flight simulators to train pilots), through operational research models to design and test new systems and tactics, to planning systems used to command and control forces. With its Defence Synthetic Environment Platform, the MoD is developing a revolutionary way of accessing and reusing SE content from a wide market of suppliers.
AT: What are the threats governments face today?
RS: Conflicts unfold at an unprecedented rate and things can change in days if not hours, the amount of data available is overwhelming in both volume and complexity. We need new tools to make sense of it all. This is against a backdrop of rising costs and financial pressures, and so there is a premium on making the most of precious resources.
The war in Ukraine has demonstrated that our adversaries are prepared to use tactics and weapons that run counter to our values. They have advanced technologies at their disposal. In simple numerical terms, we are outnumbered.
If we are to defend our people, our prosperity, our values, then we need to box clever. We need to outwit as well as outfight our enemies, we need to integrate our responses across every domain and adapt in real time. This calls for a transformation in the way armed forces train, prepare and plan.
AT: Can you tell a little about these technologies are being used?
RS: Modelling and simulation has been successfully applied to understand the interactions between limited numbers of military systems (for example, ships, tanks and aircraft) for a long time, but the effect of human behaviour on these interactions, and the reverse (the impact of military activity on civilian activity) has always been exceptionally hard to model. But it is necessary for what has been dubbed multidomain integration, and requires insights into the behaviour of highly complex systems involving large groups of people and machines operating in dynamic environments.
While AI and data science have been drivers of many recent advancements in digital technology, we see simulation as a third member of this set. All three disciplines depend on each other for progress. For example, the combination of increasing amounts of data, more capable data science techniques, and a range of machine learning techniques is offering opportunities to construct faster and more realistic simulations, able to combine observed real-life behaviour with expert judgement.
Equally, simulations are playing a crucial role in training AI. These three disciplines also come together through digital twin technology and where AI and machine learning are used to dramatically accelerate scenario exploration, planning and optimisation.
AT: Are there technologies or solutions you hope to develop, or utilise differently?
RS: One of the goals is to rapidly accelerate the process of creating SEs for a specific purpose. This currently takes years; it needs to be much faster – weeks or even days – if we are to make SEs an integral part of decision-making processes in future. Skyral includes a marketplace of content and tools to help developers rapidly compose content into SEs.
AT: You have highlighted some of today’s challenges. How can this programme address them?
RS: SEs have long been used to meet individual requirements in the defence sector. Despite the large number and considerable commonality between them, the MoD has traditionally contracted for each individual SE in isolation as a full vertical stack. The economics of doing so are rational when taken in the context of what’s best for each project, but over time this has led to a situation where MoD has paid multiple times for the same or very similar content. This means funding has focused on the needs of today, rather than the cutting-edge SE technology of the future.
Taken individually, the economics of standalone SE solutions are rational, but collectively they’re suboptimal. Skyral Exchange should enable quicker delivery of new solutions and easy reuse of existing ones. Secondly, we believe that Myridian will open the marketplace far wider, to include novel emerging content (models and data) from academics. It will also encourage SMEs to partner with academics to create novel content that would not have existed if we had waited for a requirement.
AT: Where is the UK right now in regard to these technologies?
RS: The UK punches above its weight in synthetics. We have world-leading institutions that attract leading academics and a rich seam of modellers working on theoretical models, but not that strongly linked to industry or policymakers. In some areas we have strong government modelling capability, but this is largely using outdated technology for simulation.
We need to attract industry and UK research and innovation investment into cutting edge SE technologies, proliferate this through academia and make the bar to entry extremely low for start-ups and SMEs so they can start to produce content in a cheap fashion, but still get access to new customers.
We then need to make it easy for systems integrators to be able to construct SEs rapidly. Finally, we need to support researchers in translating their foundational work to address the key technological challenges in enabling next-generation synthetics.
AT: What can this programme and these technologies ultimately offer?
RS: Planners and decision-makers in government need to respond quickly, effectively and efficiently to events that threaten our safety and prosperity. These events are often multidomain, they involve military and civilian elements, cross-border, and the data available is vast.
Bringing all of this together to track the interdependencies, explore options and then conduct a constructive debate in order to make good decisions at the speed of relevance is no longer possible without help. There is an urgent need to find solutions that support decision-makers and planners. This is not about automating decisions; humans must always have the final say. This is about accelerating our understanding of complex situations so we can have influence over them.
There is huge potential to apply SEs to challenges beyond the defence and security space. The faster and more economically we can build SEs, the more we can use them in both extreme and everyday situations to improve the way we understand and respond to complex situations. That will underpin resilience and help to address some of the most pressing issues of today and the future.
