About
We are a team of seasoned experts with over a decade of experience specializing in mobile, submarine, and drone robotics.

Our journey has taken us from working with leading companies in Italy (Magneti Marelli , SISSA , Centro di Ricerca E. Piaggio) to collaborating with top-tier tech firms in Silicon Valley (Ambarella Inc.).

Throughout the years, we have consistently delivered cutting-edge solutions, leveraging state-of-the-art technology to meet the evolving needs of our clients. With a deep commitment to innovation and collaboration, we bring a wealth of expertise and a proven track record in pioneering the future of robotics.
Past innovations

In our previous work, we have developed several foundational systems that serve as the backbone for autonomous driving and various advanced robotics tasks. Our expertise spans the entire development pipeline, from initial research and prototyping to fully integrated embedded systems.

Here are some examples of our previous works that demonstrate our expertise.

High Definition Automotive Maps
High-definition maps are essential tools for autonomous driving, requiring the acquisition and management of large amounts of data from sensors such as cameras, GPS, radar, and lidar. To process this information, we have developed tools that, thanks to advanced filtering algorithms and the extraction of semantic information, are capable of transforming raw data from low-cost sensors into useful knowledge for autonomous navigation. The result of this processing is a map that provides detailed lane-level information , ensuring precision and safety in the context of autonomous driving and ADAS systems.

Localization
The localization system developed is capable of determining the vehicle's position in the world with centimeter-level accuracy. The algorithm not only localizes the vehicle but also estimates the heading with an error of lower than one degree. The technology leverages data from low-cost sensors, such as monocular cameras, inexpensive GPS, and inertial data provided by the vehicle's control unit via CAN communication. The data are integrated, filtered and, through matching algorithms with HD maps, enable precise localization, which is essential for autonomous driving applications.

Data Fusion
The developed data fusion system allows for the identification of obstacles around the vehicle and the real-time estimation of their motion, providing an accurate perception of the surrounding environment. It uses data from sensors such as monocular cameras and radar, processing and combining the information to achieve an integrated view of the obstacles. In addition to detecting objects, it can predict their future movement, estimating possible trajectories, thereby enhancing the ego-vehicle's ability to make safe and timely decisions while driving.

Planning & Control
We have developed a complete planning and control stack for autonomous vehicles, from high-level decision-making to the generation of steering, throttle, and brake commands for trajectory tracking. The system, using HD maps and obstacle data, performs maneuvers such as lane keeping, overtaking, obstacle avoidance, and manages traffic lights, roundabouts, and highways. Built in C++ and implemented on embedded hardware, it has been successfully tested in various scenarios, including highways and urban centers, with demonstrations at CES in Las Vegas for journalists and potential customers.

Team

We are a team of scientists who have been working together for over 10 years, specializing in the research and development of advanced algorithms and software. Our strong interdisciplinary background enables us to tackle complex problems with scientific precision.

  • Gabriele Lini

    Chief Executive Officer

    He graduated with honors in Computer Engineering from the University of Parma, where he also earned a PhD in Information Technologies. He then continued his specialization at the International School for Advanced Studies (SISSA) in Trieste, where he worked on his most significant scientific paper: 'Predictable dynamics of opinion forming for networks with antagonistic interactions,' published in 2014 by IEEE Transactions on Automatic Control. He worked at Ambarella, Inc. as a Senior Manager, coordinating advanced research departments. Prior to his role at Ambarella, he spent two years at Magneti Marelli, managing a research center focused on developing software for autonomous driving.

  • Francesco Di Corato

    Chief Scientific Officer

    He graduated with honors in Robotics and Automation Engineering from the University of Pisa, where he also earned a PhD in Automation, Robotics, and Bioengineering. He has published over 40 scientific articles in leading journals in the field. He possesses strong scientific and mathematical expertise, with a focus on robust estimation and machine learning. He served as Principal Scientist at the E. Piaggio Interdepartmental Research Center, where he worked on guidance, navigation, and control systems for autonomous aerial and terrestrial vehicles. At Ambarella, Inc., he led the research team responsible for HD mapping and localization algorithms. Before joining Ambarella, he worked with Gabriele at Magneti Marelli’s autonomous vehicle research center.

  • Federico Cabassi

    Chief Technology Officer

    He graduated with honors in Computer Engineering from the University of Parma. He then pursued a PhD at the same institution, studying operational research techniques in global optimization. During this period, he participated in various projects related to robotics, automation, and process optimization, publishing several scientific articles on these topics in major international journals and conferences. For several years, he led the research group within the Path Planning department at Ambarella, Inc., overseeing the development and implementation of advanced solutions in trajectory planning and decision-making. He is also a seasoned developer with extensive experience in high-performance and embedded software development.