Robot Learning

Robots which learn how to physically interact with objects

Robot Learning Lab

The Robot Learning Lab at Imperial College London is developing the next generation of robots empowered by artificial intelligence, for assisting us all in everyday environments. Our research lies at the intersection of robotics, computer vision, and machine learning, and we are currently studying how robots can learn to physically interact with objects, using their arms and hands.

Applications include domestic robots (e.g. tidying the home), manufacturing robots (e.g. assembling products in a factory), and warehouse robots (e.g. picking and placing from/into storage). In particular, we are investigating real-world applications of deep learning for vision-based robot manipulation, using various combinations of reinforcement learning, imitation learning, and classical control.

The lab is directed by Edward Johns in the Department of Computing. To find out more about our lab research and members, and for up-to-date news, please visit our Robot Learning Lab website.