AI at Saab: Artificial eye
Saab’s growing embrace of AI technology includes the capability to obtain real-time predictions about the physical behaviours of people, ships or planes, as Joakim Ekblad explains.
In recent years we’ve seen Artificial Intelligence (AI) become a key element in much of Saab’s security and defence product portfolio, be it for the likes of surveillance sensors, smart cockpit technology or autonomous sea rescue systems.
To Saab AI is an asset that ranges from predictive maintenance to behavioural detection and it plays an important role in its future autonomous systems. Ultimately, AI is both a pathfinder and an efficiency tool, and its importance to Saab and our customers will only increase in the years to come.
What can AI give to Saab?
“AI enables us to put certain functionalities in products that we haven’t been able to do before. Things become smarter,” explains Joakim Ekblad, Head of Emerging Technologies.
“It allows us to make reasoned decisions about tens of thousands of things and solve very complex situations quickly, finding the best possible choice that’s available at the time.”
Changing how we learn and experience technology
As a trusted partner to Microsoft’s Mixed Reality programme, Saab is creating applications for various hardware platforms, including the Microsoft HoloLens, an immersive mixed reality holographic headset that has far-reaching implications for user experience and training in both civilian and military arenas.
“But it’s not just about ‘bling-bling’ applications."
“Autonomous craft and planes are often talked about but they’re still years away. What we can do today is pick the low-hanging fruits of AI, such as predictive maintenance. We have all the data that the aircraft generates; if we can use AI to find a problem with the aircraft before it breaks down we can save lots of time and lots of money by switching the parts before they break."
“This is a tremendous use-case for Saab and it’s where competitors will try to beat us. The Gripen’s life cycle is very cost effective but as AI for predictive maintenance develops further, competitors will be looking at how they can steal a march on us. So we need to step up and keep developing our capabilities and keep the life cycle cost down, otherwise they’ll beat us to it.”
New solution for massive data fusion
One example of the high-end technologies that AI can bring is Saab’s massive data fusion project.
“Our latest solution for massive data fusion is a cloud-based data lake where we take data and put it into the context of time and space,” explains Ekblad.
“This allows us to make real-time predictions about the physical behaviours of where people, ships or planes are going, and also predict the contextual behaviours of what these people, ships or planes are actually doing."
“For example, in the crowded skies above Sweden, we could use the technology to filter out planes showing normal behaviours, and focus on objects, including aircraft and drones, whose movements on radar are abnormal. We can abstract that information without needing a whole load of information on altitude, proximity to an airport, and so on, as we’ve done up till now - the AI in our massive data fusion solution can almost instantly reason to what the plane is doing.”
A new reality
Saab’s rapid integration of XM Reality’s Remote Guidance tool is an ideal example of the agile innovation culture we need to stay competitive.
The technology can also help spot unusual activities on the ground or even analyse sudden mentions in foreign media about Saab or Swedish capabilities. It has already attracted significant interest both inside and outside Saab, because it has great advantages in this age of complex, ‘grey area’ conflicts.
“Today’s wars and tensions are strange in that they can happen while civilian aircraft, road traffic or boats continue to fly, drive or sail,” says Ekblad.
“The signal environment is getting more and complex. We’re trying to respond to this on the ground but the ‘cyber weather’ is turbulent. Having the predictive power to understand changes in behavioural patterns is crucial in the long run.”
What’s next for massive data fusion, AI and Saab?
As it continues to develop in work with Saab’s sensor offerings along with space-based sensors and open source intelligence (OSINT), the massive data fusion technology will become increasingly important to Saab’s surveillance offering. Joakim Ekblad is proud of the progress he and his team, and Saab, have made from drawing board to customer contracts in just three years.
“Together with our customers for this capability we can let the company capitalise upon what we have while our team goes away and does more advanced research into AI and add to that our product,” he says.
“We also need to meet our biggest challenge: we’ve doubled the team working on this technology just over the past year, but the demand for our massive data fusion services is so big that we are looking to expand the team. Scaling up is the biggest challenge. There’s a real buzz in the company about this technology’s potential. We have exciting times ahead of us!”