Research

Interests

My main interests are in social Human-Robot Interaction (HRI), cognitive robotics and artificial intelligence. In particular, how should robots behave so they can interact naturally with and learn from people. As people are a social species, many of their cognitive abilities are rooted in social and cultural behaviour; thus robots that are socially aware and act appropriately can potentially be much more useful and effective at their tasks compared to their non-social counterparts. Studying artificial intelligence and designing human-robot interactions ultimately asks the question of what it is to be human, which I find a fascinating question to explore.

Areas of interest include: human-robot interaction, social interaction, human-robot teamwork, Search&Rescue robotics, language games, (non-linguistic) multi-modal communication, computer vision, evolutionary roboticscognitive systems and philosophy.

 

Present

I am currently working as a postdoc at Delft University of Technology in the Interactive Intelligence Group on the TRADR project. The TRADR project is about creating robots that can help during search and rescue missions, such as fires, earthquakes or other (natural) disasters. The aim is to create robots that work alongside rescue workers as team-mates, rather than as tools.

 

Past

I have been employed by Plymouth University as a research fellow, working on the ALIZ-e project which aims at developing social robots for prolonged interaction with children with diabetes. By enabling robots to interact naturally with children, they can fulfil mentor and buddy roles, and may provide support and motivation.

I did my PhD research at Plymouth University in the United Kingdom, under the supervision of dr. Tony Belpaeme. I was involved in the CONCEPT Project together with Frédéric Delaunay. My main topic of interest was the learning of concepts in cognitive system through social interaction.

The ability to use concepts lies at the core of human intelligence. To create this capacity in artificial intelligent systems a thorough understanding is needed of how humans are able to learn and apply conceptual knowledge. To date the topic of concepts (or categories, classes) is studied in a wide variety of disciplines, but no comprehensive, all-inclusive theory has been developed yet.

Relevant code can be found here and here.

 

Reviewing
I have been doing regular reviews for HRI, ICSR, RO-MAN, ICDL-EpiRob, IROS, ICRA and IJSR.