Research activities





About this site

About me

I am passionate about projects at the interface between intelligent embedded systems and large scale data science and in particular projects related to machine perception.

I am an expert in wearable computing and computational behavioral analytics - the art of understanding human behavior by the right combination of sensors, signal processing, machine learning and AI techniques. However, I also have also a strong interest in mobile robotics and robot perception, the field in which I carried out my PhD.

I am an engineer and a scientist.

As an engineer, I enjoy hands-on embedded systems and data science projects. I have a profound understanding of hardware-software co-design and the trade-offs involved in designing real-time signal processing and machine learning systems. An example of this is BlueSense - my Wearable Motion Sensing, IoT and Sensor Research Platform.

As a scientist, I lead a group of PhD students and postdoctoral researchers to advance computational behavioural analytics, with a particular focus on AI and machine learning for human activity recognition and new sensors technologies.

As of 2018, I obtained £3.2m total research funding from the EU, the Swiss NSF, the UK EPSRC, the Austrian FFG, Huawei Technologies, Unilever, and through a Google Faculty Research Award.

Some of my work on deep model for activity recognition is cited in an Apple technical document and many of my datasets have become references in the community.

I have managed large multi-site interdisciplinary research projects, such as the 1.5m€ EU-FP7 project OPPORTUNITY (2009-2012) involving 10+ PhD students across 4 sites.

I have applied my research to pervasive healthcare, manufacturing, sports, HCI, crowd behaviour, and others. I organised several machine learning challenges, such as the 2013 OPPORTUNITY Gesture Recognition Challenge, or the 2018 Sussex-Huawei Locomotion Recognition Challenge.

I obtained a PhD from EPFL where where I graduated in 2005. During my PhD I developed bio-inspired electronic circuits with fault-tolerance, learning, and developmental capabilities that were applied to the control of autonomous mobile robots. This work was carried out in context of the EU FP5 FET project POEtic.