Dstl funds artificial intelligence for future Royal Navy operations

The system is designed after the human brain, emulating the mammalian conditioned-fear response mechanism

Defence Science and Technology Laboratory (Dstl) has awarded a contract of a £1 million investment to Roke Manor Research, an electronics engineering consultancy, to develop artificial intelligence software that is designed to help the Royal Navy handle the growing complexity of threats.

Named STARTLE, the machine situational awareness software continuously monitors and evaluates potential threats using a combination of artificial intelligence techniques. It is designed after the human brain, emulating the mammalian conditioned-fear response mechanism.

Roke will demonstrate the STARTLE techniques by integrating them into the Open Architecture Combat System, a demonstrator designed to show the utility of research ideas in a representative combat system in a realistic environment.

If integrated into existing warship sensor suites, it would support the principal warfare officer by processing multiple sources of information, while cueing systems to assess and confirm potential threats. In increasingly complex and dynamic mission environments, it could allow the command team to make better informed decisions faster. 

Mike Hook, lead software architect on STARTLE at Roke, said: “This is an exciting project for us. Traditional methods of processing data can be inefficient so we have looked at the human brain’s tried and tested means of detecting and assessing threats to help us design a better way to do it. The techniques have the potential to benefit the Royal Navy.

“The first two phases of the project have proven that we’ve been able to successfully apply these techniques to real data from complex scenarios. The clever part comes in the way these potential threats are detected and the way our software redistributes resources to decide if they are real – all in the blink of an eye.”

In addition to maritime defence systems, STARTLE can also be adapted for autonomous vehicles, and health and usage monitoring applications.