Dimitrios Kanoulas
University College London
Dimitrios Kanoulas is an Associate Professor who received his PhD degree from Northeastern University, Boston. He was a Postdoctoral Researcher at the Italian Institute of Technology for five years. He is currently an Associate Professor in Robotics with the Department of Computer Science, University College London, and a UKRI Future Leaders Fellow. He has published over 60 research papers in high-impact robotic journals and conferences. His research aims to apply perception and learning in robotics, with a focus in legged ones. He is developing new cognition and planning algorithms for articulated robots with a high number of Degrees-of-Freedom, that locomote and manipulate in uncertain natural environments. This includes real-time methods (geometric-based or learning-based) for sensing, mapping, and self/environment interaction modeling. Self-supervised, deep, reinforcement, and imitation, learning, as well as sim2real and robot teleoperation, belong to his recently explored areas of interest and are connected to the ongoing funded projects. Experimental justification of theory on real robots plays a central role in his research and academic philosophy. To date, he has developed perception and learning techniques on mini-bipeds, half-size and full-size humanoid robots, as well as on several quadrupedal robots and mobile manipulators, facing all the big challenges of real-world scenarios. He has received the Best Interactive Paper Award at Humanoids 2017, Best Student Paper Finalist at ICARCV 2018, the Marie Sklodowska-Curie Actions Seal of Excellence 2018, the Best Associate Editor Award at ICRA 2022, and Outstanding Paper Finalist at ICML (WFVML workshop) 2023. He has also received the UCL-CS Early Career Researcher of the Year 2022 and the UK-RAS Early Career Award 2022.