Combining multiple scientific education topics could help unlock the next generation of robotics, according to a new report from Imperial College London.
Published in Nature Machine Intelligence, the report suggests combining materials science, mechanical engineering, computer science, biology and chemistry into a unified teaching discipline. Students qualifying with this intersection of skills could be the driving force behind what the researchers term ‘Physical AI’, where the software and hardware of the future dovetail in the form of intelligent, life-like robotics.
“The development of robot ‘bodies’ has significantly lagged behind the development of robot ‘brains’,” said co-lead author Professor Mirko Kovac from Imperial’s Department of Aeronautics and Switzerland’s Materials and Technology Centre of Robotics.
“Unlike digital AI, which has been intensively explored in the last few decades, breathing physical intelligence into them has remained comparatively unexplored.”
According to the Imperial team, the reason for this gap is that education has yet to come up with a coherent system for combining the topics required to make Physical AI a reality. Equipping tomorrow’s scientists and engineers with a broader range of complementary skills would help address that gap and could lead to the development of lifelike robots with biomimetic capabilities, such as adaptable body control, autonomy and real-time sensing.
“The notion of AI is often confined to computers, smartphones and data-intensive computation,” Kovac continued.
“We are proposing to think of AI in a broader sense and co-develop physical morphologies, learning systems, embedded sensors, fluid logic and integrated actuation. This Physical AI is the new frontier in robotics research and will have major impact in the decades to come, and co-evolving students’ skills in an integrative and multidisciplinary way could unlock some key ideas for students and researchers alike.”
According to the researchers, achieving biomimetic functionality in robots will not only require combining conventional robotics with AI and other disciplines, but also deeper interdisciplinary collaboration across a wide range of sectors.
“We envision Physical AI robots being evolved and grown in the lab by using a variety of unconventional materials and research methods,” said Kovac. “Researchers will need a much broader stock of skills for building lifelike robots. Cross-disciplinary collaborations and partnerships will be very important.”