Engineers at the University of Pennsylvania claimed to have developed an algorithm for giving a robot real-time object control.

Footage shows a machine powered by the technology designers called Consensus Complementarity Control (C3) exhibiting complex physical interactions as it slides a tray off a platform and returns it.

They said the innovation could revolutionise how robots engage with their environment by tackling previously deemed too intricate tasks, such as manipulating sliding objects.

Assistant Professor in the Mechanical Engineering and Applied Mechanics department Michael Posa said his Dynamic Autonomy and Intelligent Robotics Lab team designed C3 to translate high-level directives from AI, like chopping an onion, into precise physical actions.

He said: 'This kind of lower and mid-level reasoning is really fundamental in getting anything to work in the physical world.'

The Professor added the algorithm's breakthrough lies in its ability to simulate different contact scenarios, allowing robots to make informed decisions on the fly about how to interact with objects.

William Yang, a recent doctoral graduate, showcased a robot using C3 to control a tray, a task involving complex dynamics like sliding, which is particularly difficult to manage in robotics.

Yang's work, which earned the Outstanding Student Paper Award at the 2024 Robotics: Science and Systems conference, marks a significant step forward.

He said: 'Sliding is notoriously hard to control in robotics.'

The innovation claimed that the C3 algorithm would make robots not just smarter but significantly more capable of handling the unpredictable nature of the physical world.

The team said they aim to refine C3 to handle more varied and dynamic environments.