Studying human biology will form the basis of research into robotics at a new centre in the University of Ulster

A new research centre in the University of Ulster is aiming to integrate sophisticated computer systems with basic human biology, with a view to making breakthroughs in the fields of intelligent systems, artificial intelligence and robotics.

The new £20 million (€29.4 million) Intelligent Systems Research Centre (ISRC) represents the biggest single research investment in the 150-year history of the University's Magee campus. Invest Northern Ireland has offered the university £7.2 million (€10.6 million) of funding, £1.1 million (€1.6 million) of which is from the Centres of Excellence programme, with a further £6.1 million (€8.9 million) from the Integrated Development Fund, co-ordinated by Ilex, the urban regeneration company for the city.

The concept of intelligent systems is based on the theory that making computer systems smarter requires knowledge of biology and the best examples of intelligent systems on the planet, explains Prof Martin McGinnity, director of the new centre.

"Obviously the most intelligent biological systems are humans and we need to see how humans process information," says Prof McGinnity.

Knowing how areas of human function work, such as visual capability, hearing and tactile senses, means that they can then be incorporated into computer systems, he says.

Although these sophisticated systems could be used in a normal desktop PC, they are more likely to be found in other environments, explains McGinnity. "These machines would be used in advanced manufacturing, entertainment systems or new intelligent transportation systems," he says.

The projects ongoing in the ISRC are a result of collaborations between computer scientists, electronic engineers, neuroscientists and psychologists, says McGinnity.

"When you work out how biology does it as best you can and then extract the principle behind it and try and put that into computer systems."

According to McGinnity, while modern computer systems are very sophisticated they only do what they have been pre-programmed to do.

Mobile robots would encounter all sorts of scenarios but it would be impossible to envision them all, says McGinnity. "We are trying to make systems that have some of the same characteristics as humans with regards to intelligence, for example the ability to perceive, learn, reason and adapt to unanticipated circumstances."

Researchers at the center are working on the creation of large scale neuron networks, involving millions of neurons that will mimic the human brain to some extent, says McGinnity.

One area of application of cognitive robotic systems would include service robotics in the home, for example robots that assist the elderly or the disabled. "The increasing proportion of elderly people in the population is placing a rising burden on the healthcare system and some of these techniques might actually assist that," explains McGinnity.

Wireless ambient intelligence can also be used to create a home environment where it will be possible to monitor people's health unobtrusively, says McGinnity.

"It could give advance warning of problems that could arise. Intelligent systems that can detect when someone has fallen or is about to fall would be of huge value," he says.

According to McGinnity, much of the work the ISRC do is theoretical but there are also strong links with industry, with a view to creating tangible products and services.

"We bring intelligent techniques to industry, such as improving process control, minimising scrap or creating smarter more autonomous machines. All these things help improve yield and enhance efficiency and profitability."

A business development manager will look after the commercialisation of the research output of the centre, he adds.

The multi-million pound centre is a major expansion of the existing Intelligent Systems research group in the school of computing and intelligent systems and will allow the group to broaden their research interests considerably, says McGinnity.

"This will substantially expand our capability and we'll be able to work in areas that we previously were unable to due to lack of resources," he says.

INTERWAVE

Interwave stands for Interconnections by Electromagnetic Wave Propagation in Silicon-based Artificial Spiking Neural Networks.

Researchers are working to develop a completely radical approach to the implementation of high density spiking (information-sending) neural networks.

The potential of this approach is enormous, as it allows - for the first time ever - the possibility of unlimited scaling of neural networks in hardware.

SENSEMAKER

This project aims to conceive and implement electronic architectures that are able to merge sensory information sampled through different modalities into a unified perceptual representation of our environment.

The ultimate ambition is to create new senses to supplement the existing senses, for example, to allow one to see the thunder or hear the lightning.