Technology to improve safety in the ‘danger zone’
Engineers at the University of Southampton will unveil new technologies to improve communications in ‘danger zones’ at a symposium in Farnborough next week.
The outcomes of the ALADDIN research programme (Autonomous Learning Agents for Decentralised Data and Information Networks), will be presented at a final research symposium at the Park Centre in Farnborough on Monday 27 September.
ALADDIN is a multi-million pound multidisciplinary research project led by Professor Nick Jennings from the University of Southampton’s School of Electronics and Computer Science (ECS), who heads the world’s largest agents research group, and funded by BAE Systems and the Engineering and Physical Sciences Research Council (EPSRC). It involves research groups from the Universities of Bristol, Oxford, and Imperial College London.
Focusing on developing techniques and technology to overcome the challenges facing different agencies involved in an emergency response, ALADDIN is considered a best-practice project by the EPSRC.
Over five years of research, the ALADDIN partnership has developed new techniques for building decentralised autonomous systems in highly uncertain dynamic environments. It has made it possible to deploy systems of autonomous agents which can plan and reason and pass information to one another without direct human control.
During the research project, the partnership developed three demonstrators which show how ALADDIN algorithms work in disaster response situations:
- A situational awareness demonstrator involving weather sensors.
- A disaster rescue simulation, considering how agents could operate and communicate with one another in a disaster scenario like an earthquake.
- A building evacuation simulator, involving the use of autonomous agents in a disaster scenario in a tower block or on a boat. The agent’s job here is to get people to safety.
“We have done some really groundbreaking work on the development and use of autonomous agents in decentralised systems,” says Simon Case, BAE Systems. “The challenge in practice is to detect whether the information coming from each agent is accurate or to be able to ascertain if it is not reliable.”
ALADDIN outputs have already been used in various customer communities, including weapon assignments, resource management in the Unmanned Aircraft System (UAV) sector, and more recently, for an instinct programme for Detica and the Home Office.