Dimitri Ognibene

Dimitri Ognibene, PhDDimitri Ognibene
Post Doctoral Research Associate

Email: dimitri.ognibene@kcl.ac.uk


View Dimitri Ognibene's profile on LinkedIn View Dimitri Ognibene's profile on Research Gate View Dimitri Ognibene's profile on Google Scholar View Dimitri Ognibene's Homepage


Dr. Dimitri Ognibene is Research Associate in the Thrish Lab at Department of Informatics of King’s College London. He was research associate at Imperial College London in the Personal Robotics Laboratory led by Dr. Yiannis Demiris. He obtained his Ph.D. in Robotics from the University of Genoa and worked in Rome at the Institute for Sciences and Technologies of Cognition of the National Council of Research. He is interested in robotics, artificial intelligence, software engineering and cognitive neuroscience. He has worked on active vision models for dynamic and social environments, adaptive models of active perception based on reinforcement learning, bio-inspired motor control primitives and proactive allocation of resources in artificial and biological agents.

Recent work: Active perception in dynamic and social environments [1,2]

[Work developed at Imperial College Personal Robotics Laboratory with Dr. Yiannis Demiris]

The video shows the robot while predicting human actions and exploring the environment to find the targets of human actions. To understand what is happening around it and produce effective response, the robot has to actively find the most relevant elements of the environment, e.g. action target, exploiting the cues that it has already discovered, like the movement of the human hand.


[Work developed at Imperial College Personal Robotics Laboratory with Dr. Kyuhwa Lee and Dr. Yiannis Demiris]

The video shows the robot while interacting with a human and collaboratively mixing music. The robot interacts proactively. It  predicts partner’s actions based on his knowledge on the “music style”. It uses its prediction to anticipatively direct attention to perceive human actions. Then the robot  correct its previous predictions based on the actually observed human actions and finally selects its next action.


Directing robot attention to recognise activities and to anticipate events like goal-directed actions is a crucial skill for human-robot interaction. Unfortunately, issues like intrinsic time constraints, the spatially distributed nature of the entailed information sources, and the existence of a multitude of unobservable states affecting the system, like latent intentions, have long rendered achievement of such skills a rather elusive goal. The problem tests the limits of current active perception systems. It requires an integrated solution for tracking, exploration and recognition, which traditionally have been seen as separate problems in active vision.

Selected Publications

  1. Ognibene, D., Chinellato, E., Sarabia, M., Demiris, Y. Contextual action recognition and target localization with active allocation of attention on a humanoid robot, Bioinspiration & Biomimetics 8(3), 2013

  2. Ognibene, D., Demiris, Y. Toward active event recognition, Proceedings of the 23rd International Joint Conference on Artificial Ingelligence (IJCAI), Bejing, China, Aug 2013

  3. Pezzulo, G., Ognibene, D. Proactive Action Preparation: Seeing Action Preparation as a Continuous and Proactive Process, Motor Control 16(3):386-424, 2012

  4. Ognibene, D., Pezzulo, G., Baldassarre, G. How can bottom-up information shape learning of top-down attention control skills?, Proceedings of the 9th International Conference on Development and Learning (ICDL), Ann Arbor, Michigan USA, Aug 2010


Recent Publications

  1. Ognibene, D., Baldassarre, G. Four problems of active vision: bio-inspired solutions from integrating bottom-up and adaptive top-down attention in a camera-arm robot, accepted, IEEE Transactions in Autonomous Mental Development

  2. Nivel, E., Th ́orisson, K.R., Ognibene, D., Steunebrink, B., Dindo, H., Pezzulo, G., Rodriguez, M., Corbato, C., Schmidhuber, J., Sanz, R., Helgason, H.P., Chella, A. AUTONOMOUS ACQUISITION OF NATURAL LANGUAGE, accepted for Intelligent Systems and Agents 2014

  3. Ognibene, D., Catenacci Volpi, N., Pezzulo, N., Baldassarre, G. Learning Epistemic Actions in Model-Free Memory-Free Reinforcement Learning: experiments with a neuro- robotic model, Proceedings of Living Machines, London, UK, Jul 2013