I'm an engineer and developer with knowledge sharing and problem solving attitudes.
I have 6 years experience in recursive Bayesian filtering and software developing, including R&D collaboration with Selex ES (Finmeccanica) and a Ph.D. in computer science and automation engineering from University of Florence (Italy). I was recipient of the 2014 Australia Award Endeavour Research Fellowship (granted by the Australian Government, Department of Education) under which I have been an R&D collaborator at the Advanced Signal Processing Group of Curtin University of Technology (Perth, Australia).
From January 2016, I'm a post-doc at the Istituto Italiano di Tecnologia (Genoa, Italy) in the iCub Facility, Humanoid Sensing and Perception, for studying, researching and developing new augmented reality-based visual tracking systems and visual servoing controllers for the humanoid robotic platform iCub.
I actively contribute to the development of the free and open source
iCub software as a member of the
I also contribute to and develop other open source project during my research activities, which can be found in my
This work addresses recursive markerless estimation of a robot’s end-effector using visual observations from its cameras. We formulate the problem into the Bayesian framework and address it using Sequential Monte Carlo (SMC) filtering (also known as particle filtering). We use a 3D rendering engine and Computer Aided Design (CAD) schematics of the robot to virtually create images from the robot’s camera viewpoints (likewise in augmented reality contexts). These images are then used to extract information and estimate directly the 6D pose of the end-effector. To this aim, we developed a particle filter for estimating the position and orientation of the robot’s end-effector using the Histogram of Oriented Gradient (HOG) descriptors to capture robust characteristic features of shapes in both cameras and rendered images. We implemented the algorithm on the iCub humanoid robot and employed it in a closed-loop reaching scenario. We demonstrate that the tracking is robust to clutter, allows compensating for errors in the robot kinematics and servoing the arm in closed loop using vision.
To precisely reach for an object with a humanoid robot, it is of central importance to have good knowledge of both end-effector, object pose and shape. In this work we propose a framework for markerless visual servoing on unknown objects, which is divided in four main parts:
The pipeline prove to be effective and robust and has been tested on the iCub humanoid robot platform, achieving real-time computation, smooth trajectories and subpixel precisions.
The Ph.D. thesis "Distributed multi-object tracking over sensor networks: a random finite set approach" has been evaluated to be a significant scientific contribution by the Florence University Press (FUP) judging committee.
Issued by the Department of Industry of the Australian Government. The Endeavour Research Fellowship provides financial support for postgraduate students and postdoctoral fellows to undertake short-term research (up to 6 months) towards a Masters or Ph.D. or postdoctoral research in any field of study in Australia.