About me

I am currently a Research Assistant at the Robotics and Perception Group working on event-based vision for robotics, advised by Prof. Davide Scaramuzza. I have successfully completed my MSc in Robotics, Systems, and Control at ETH Zürich in October 2023. I am currently seeking admission into a PhD program to further advance my academic career and contribute to cutting-edge research in the areas of computer vision, deep learning and state estimation.

I have recently completed an internship at NASA Jet Propulsion Laboratory, where I had the chance to work on the vision system for state estimation of the Mars Science Helicopter. During my Master's Degree, I have gained quite some research experience by working on different projects in different areas of Robotics and AI. Prior to my MSc at ETH in Robotics, I obtained a Bachelor's Degree in Automation Engineering at Alma Mater Studiorum - University of Bologna in 2020.

Publications

Structure-Invariant Visual-Inertial-Odometry

Supervised by Jeff Delaune, Giovanni Cioffi and Davide Scaramuzza

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2024 (Oral presentation)

Webpage | Paper | Video

A novel range-VIO method that fuses altimeter measurements with a visual-odometry framework, and eliminates the need for any type of ground planarity assumption.

News

04/2022	I joined Verity AG in Zürich as a System and Estimation engineer
09/2021	I was TA at the Rehabilitation Engineering Lab (ETH Zürich) for the course of Physical Human Robot Interaction
07/2021	I took part in the Robotics Summer School organized by the RobotX initiative at ETH Zürich
07/2020	I completed my BSc studies after defending my Thesis named Defect Detection in the Tyre Manufacturing Process using Convolutional Neural Networks (project in partnership with Pirelli Tyres S.p.A.)

Projects

	Instinctive Robot Control via Hololens2 Team members: Ivan Alberico, Michael Baumgartner, Jonathan Becker, Seif Ismail Webpage \| Video \| Report \| Code Mixed reality interface on a Microsoft Hololens 2, with which the user is able to remotely control a robotic arm and perform basic assembly tasks using hand and eye tracking.
	Monocular Markerless 6D Pose Estimation of ANYmal Team members: Ivan Alberico, Kexin Shi Supervised by Jonas Frey, Victor Klemm and Marco Hutter Webpage \| Poster \| Video \| Report An accurate toolbox for localization and pose estimation of ANYmal without external sources like depth cameras or QR codes.
	Learning to Generate Events using Spiking Neural Networks Supervised by Daniel Gehrig, and Davide Scaramuzza Webpage \| Report A learning-based solution that converts any existing video dataset recorded with conventional cameras to synthetic event data.
	End-2-End Self-Supervised SLAM Team members: Ivan Alberico, Mian Akbar Shah, Selim Kaelin, Emilk Sempertegui Webpage \| Report \| Code An online self-supervised SLAM pipeline for real-time dense reconstruction, with an online adaption module to boost performance on indoor scenes.
	Multi-task Learning for Autonomous Driving Team members: Ivan Alberico, Nicola Loi Webpage \| Github Multi-task Learning for semantic segmentation, depth estimation and 3D object detection for autonomous driving scenes.
	Planning and Decision Making for Autonomous Robots Team members: Ivan Alberico, Nicola Loi Webpage \| Github Implementation of a path planning pipeline in a dynamic environment. Global path planning, local path planning, and obstacle avoidance for navigating a spacecraft to a goal region while avoiding static and dynamic asteroids.
	Skater Blob game on Unity Team members: Ivan Alberico, Adrian Hartmann Webpage \| Github Implementation of a skateboarding game using Unity3D and Blender. Every aspect, from crafting the main 3D models to animating characters and designing the game's logic and levels, was meticulously implemented from scratch.
	Admittance Controller on Haptic Paddle Team members: Ivan Alberico, Giulio Schiavi Webpage \| Github Implementation of an admittance controller with inner position/velocity loop on an haptic device, with the aid of a servo-amplifier and tachometer for motor control.

Personal

I love playing the piano, photography, cooking and traveling. Check out my Instagram profile for more pictures. I have practiced karate for 11 years achieving black belt. My hobbies can be summarized by the following slide:

Provehito in Altum.

	Instinctive Robot Control via Hololens2 Team members: Ivan Alberico, Michael Baumgartner, Jonathan Becker, Seif Ismail Webpage \| Video \| Report \| Code Mixed reality interface on a Microsoft Hololens 2, with which the user is able to remotely control a robotic arm and perform basic assembly tasks using hand and eye tracking.
	Monocular Markerless 6D Pose Estimation of ANYmal Team members: Ivan Alberico, Kexin Shi Supervised by Jonas Frey, Victor Klemm and Marco Hutter Webpage \| Poster \| Video \| Report An accurate toolbox for localization and pose estimation of ANYmal without external sources like depth cameras or QR codes.
	Learning to Generate Events using Spiking Neural Networks Supervised by Daniel Gehrig, and Davide Scaramuzza Webpage \| Report A learning-based solution that converts any existing video dataset recorded with conventional cameras to synthetic event data.
	End-2-End Self-Supervised SLAM Team members: Ivan Alberico, Mian Akbar Shah, Selim Kaelin, Emilk Sempertegui Webpage \| Report \| Code An online self-supervised SLAM pipeline for real-time dense reconstruction, with an online adaption module to boost performance on indoor scenes.
	Multi-task Learning for Autonomous Driving Team members: Ivan Alberico, Nicola Loi Webpage \| Github Multi-task Learning for semantic segmentation, depth estimation and 3D object detection for autonomous driving scenes.
	Planning and Decision Making for Autonomous Robots Team members: Ivan Alberico, Nicola Loi Webpage \| Github Implementation of a path planning pipeline in a dynamic environment. Global path planning, local path planning, and obstacle avoidance for navigating a spacecraft to a goal region while avoiding static and dynamic asteroids.
	Skater Blob game on Unity Team members: Ivan Alberico, Adrian Hartmann Webpage \| Github Implementation of a skateboarding game using Unity3D and Blender. Every aspect, from crafting the main 3D models to animating characters and designing the game's logic and levels, was meticulously implemented from scratch.
	Admittance Controller on Haptic Paddle Team members: Ivan Alberico, Giulio Schiavi Webpage \| Github Implementation of an admittance controller with inner position/velocity loop on an haptic device, with the aid of a servo-amplifier and tachometer for motor control.