Since my early studies, I have always been fascinated by how a complex system like the human body can be regulated and controlled by an even more complex system, the brain. Specifically, understanding how a movement is conceived, planned and controlled by the brain is one of the biggest neuroscientific challenge that attracted my interest. Transposing the movement control mechanisms in a model for artefacts is an extraordinary way to tackle this challenge, to extend the knowledge on the brain and develop better robots.

In particular, it was through my double PhD degree that I could explore all the potential of neuroscience knowledge applied to robotics, and vice versa.

Through my studies at the College de France I developed strong expertise in cognitive neuroscience, specifically in the motor control. This has led me to run experiments on locomotion on healthy volunteers, and to start a thriving collaboration with neuroscientists. This allowed me to understand the main principles that regulates human locomotion, such as the head stabilization and the sense of movement. 

At the Biorobotics Insitute, I integrated this acquired expertise, deriving principles related to gaze control and providing models for robot control. In particular I developed models of eye movements (saccade, smooth pursuit and vestibulo-ocular reflex) and integrate this with head movements. I was able to develop a complete model of gaze stabilization including vestibulo-collic, vestibulo-ocular and optokinectic reflex.

I have been working on different projects, in particular RoboSoM, where I was responsible for the integration of the previously mentioned gaze models in a humanoid robot performing locomotion tasks.

In 2013, I joined the Human Brain Project, the EU flagship initiative on the brain, working in the SubProject 10, dedicated to Neurorobotics. I started as Post-doc and team leader in a group of two PhD students and a research assistant. In the framework of this project I mainly contributed to the design and development of the Neurorobotics Platform (Falotico et al., 2017). This platform is a unique tool that integrates robot and brain simulators, where neuroscientists and neuroroboticists can test brain models at different levels (from point-neuron models to more abstract algorithms). In the framework of this project I have also developed models of specific areas of the brain, like cerebellum and spinal cord, tested on robotic tasks.

In 2018, I was asked to take on the role of SP10 co-leader. This had led to an increase of scientific and management responsibilities, being both main Principal investigator at the BioRobotics institute and one of the two representatives of SP10 in the Science and Infrastructure Board (SIB).

Since 2017, by relying on my expertise in AI and computational neuroscience, I contributed to the development of control models for soft robots. Soft robots, that are complex robotic platforms difficult to model, benefit from  model-free approaches. I contributed to the design and implementation of several models based on learning the kinematics and dynamics through neural networks. I have become also very active in the soft robotics community through the participation in workshops and the role of publicity chair in the first IEEE RoboSoft conference held in Livorno in 2018. I am now PI for SSSA of the PROBOSCIS project whose aim to develop a robot inspired by the elephant trunk.