The £1 million research project, funded by the Leverhulme Trust under the Research Leadership Award scheme, will be led by Dr Nathan Lepora from the University of Bristol and Bristol Robotics Laboratory (BRL). The five-year research program on ‘A biomimetic forebrain for robot touch’ will apply leading neuroscience, psychology and philosophy of the brain to develop a biomimetic forebrain embodied on a 3D-printed robot hand, aiming for human-like tactile dexterity.
The development of effective robot hands with a sense of touch is crucial for future technology: without human-like tactile dexterity, robots will not be able to solve the interactive tasks predicted for advanced manufacturing, such as autonomous assembly lines; assisted living, including personal robot helpers; food production, for example picking and sorting robots; and healthcare, such as nursing and surgical robots. Robots currently cannot do these tasks because no one knows how to combine what the hands are feels and how they control their action. Yet clearly, this problem has been solved in the human brain and hand.
Dr Nathan Lepora, Senior Lecturer in Robotics at the University of Bristol and BRL and leader of the Tactile Robotics research group said: “People see a robotics revolution happening but many things won’t be achieved if robots don’t have hands that they can use to dexterously control the world around them through a sense of touch.
“It’s about bringing together tactile hands and algorithms based on how the brain works. Why is it so difficult? The human hand has evolved over tens of millions of years. People’s intelligence has evolved in tandem with their hands to give them their unique role in the animal kingdom of being able to manipulate their surroundings so profoundly. So replicating that in a robotic device is both a challenge and an inspiration.”
The Tactile Robotics research group at Bristol Robotics Laboratory researches aspects of tactile sensing from hardware design to biomimetic and AI algorithms for perception and control. The group’s recent work on an open-source 3D-printed fingertip, which can ‘feel’ in a similar way to the human sense of touch, won Harvard University‘s international Soft Robotics competition for its contribution to soft robotics research.