The work is part of the pan-European Cognitive Control Framework for Robotic Systems (CORBYS) project that aims to produce robots that can accurately predict human behaviour.
Over the next four years, researchers at Hertfordshire University will spend €780,800 (£672,600) developing two prototypes that they hope will show a certain level of ‘robotic intuition’.
The first will be a mobile robot-assisted gait rehabilitation system that is capable of matching the requirements of patients to different stages of treatment.
Currently, robots that help people walk need constant monitoring from therapists. The natural tendency of patients is to offload effort onto the machine, which can delay recovery.
‘What you need is something that understands what’s necessary to help, but also understands how far it needs to challenge the human to develop their own capabilities,’ said lead researcher Dr Daniel Polani.
‘Think of it a bit like a dance. Sometimes if the lead dancer doesn’t lead properly, then the lead is overtaken by his partner. And that can work, but the partner also needs to know when to step back.’
Polani and his team will primarily use kinetic information, supplemented by visual data and brain activity, to develop algorithms that will allow the system to accurately anticipate the patient’s movements.
A further prototype will demonstrate robotic assistance to humans in exploratory and potentially dangerous environments. Polani believes that their research could lead to the development of a machine with high levels of intuition within the next 20 years.
‘We are pretty much on uncharted territory here,’ he said. ‘But the bigger picture in human and robotic interaction is, for us, very exciting.’
Red Bull makes hydrogen fuel cell play with AVL
Formula 1 is an anachronistic anomaly where its only cutting edge is in engine development. The rules prohibit any real innovation and there would be...