January 20, 2016      

As researchers embark on an innovative project to introduce robots that can understand human emotions in Scottish schools, there are growing signs that the country is developing a nascent artificial intelligence sector. So, what are the prospects for the emergence of Scotland as a global AI hub?

Researchers at Heriot-Watt University in Edinburgh are key partners in the EU-funded EMOTE project, which is investigating and developing a socially aware and empathic robot tutor that will use information about the emotional state of the user to improve its ability to support and motivate students.

Trials of the robots occurred at a high school in Midlothian last autumn. Ultimately, the team hopes that schools across Scotland will invest in the devices.

Project co-leader Ruth Aylett, a professor of computer science at Heriot-Watt University, is working alongside researchers in Germany, Portugal, and Sweden to develop two application domains for 11-14 year olds using an Aldebaran Nao robot and a multi-touch table.

“One involves a single user learning skills related to maps, and the other involves two users playing a sustainable energy game on the table with the robot,” she said. “Heriot-Watt has worked most on the map skills demonstrator, as well as developing a handheld version of it for field trips.”

‘World-leading research base’

The school project comes in the wake of several other interesting Scottish-based robotics and AI initiatives.

Demonstration of EMOTE researchLast October, Heriot-Watt University and the University of Edinburgh were jointly awarded £6 million ($8.52 million) by the British government to further develop the Edinburgh Centre for Robotics into a “world-leading research base in robotics and autonomous systems.”

In addition, the North Sea oil and gas sector continues to develop autonomous underwater inspection vehicles, and the new Queen Elizabeth University Hospital in Glasgow has recently introduced a £1.3 million ($1.85 million) fleet of 26 robots to move medical equipment.

According to Dr. Subramanian Ramamoorthy, a reader in the School of Informatics at the University of Edinburgh and executive committee member of the Edinburgh Centre for Robotics, there are currently “many exciting projects within the Scottish AI sector,” which he said “punches well above its weight.”

“Edinburgh has a long tradition and history of contributing at the forefront of artificial intelligence, going all the way back to the early years of AI,” he said.

The University of Edinburgh has “leading expertise” in several areas, including computer vision, machine learning, and natural language technologies said Ramamoorthy.

He pointed out that other Scottish universities are also active in the use of machine-learning methods for medical imaging and medical devices such as prosthetics.

Ramamoorthy said his own work is motivated by the desire to create robots that can learn to quickly interact with people in a natural and context-aware manner.

Scotland is ‘well equipped’

Aylett agreed that Scotland is “very well equipped indeed” when it comes to higher education expertise. She also said that Scottish universities have specific strengths in the development of big data applications and pointed out that government funding through the Scottish Informatics and Computer Science Alliance (SICSA) has helped to “bring together a critical cooperating mass” in this area.

“One example is the Heriot-Watt Underwater Robotics Group, which is developing AUVs [autonomous underwater vehicles] for pipeline and platform inspection and has spun off the successful Seebyte,” Aylett said.

More on AI and U.K. Robotics:

“In addition, the computer games industry and multimedia service industries are also potential domains of application,” she noted.

However, on a more cautious note, Ramamoorthy warned that there is currently too much emphasis on “immediate translation” to commercial products, which he argued may result in “less attention and support for fundamental issues” that continues to be necessary if the sector is to “ultimately solve the very big problems in AI.”

“It is important to continue to support fundamental research in these areas, even when their translation might take longer,” he said.