Our task is to develop new-generation underwater and manipulation technology for tomorrow’s Autonomous Robots

Our future

Despite the remarkable progress made in Robotics and AI, a number of bottlenecks may well emerge in these fields in the near future.

3D printing and design

3D printing systems are becoming more widespread as they are extremely user-friendly and affordable.

Non-conventional algorithms

Controlling robot manipulators in the real world is a highly challenging task due to uncertainty in the system's state estimation.

Marine robotics manipulation
Best Solutions. Zero Compromises.

Very low-cost 3d printed arm
Merges rigid and compliant parts (as our own arms’ bone and muscles!)
Carefully Engineered for underwater utilization by real Marine Robotics experts
BioInspired Deep Reinforcement Learning control (and that’s just the beginning!)


We have a three-year plan to develop new methodologies and low-cost prototypes and techniques for navigation and efficient underwater manipulation.
Technology Transfer

Strategic elements of systems with high technology transfer to promove technological transfer and innovation in real contexts for both industrial and societal needs

Affordable products

We develop low-cost prototypes for demonstration purposes, paving the way for actual production and deployment of real-life robots

Innovation and development

efficient techniques for solving underwater manipulation problems, leading to innovation and technological progress.

The underwater environment

is an extremely challenging scenario featuring radipidly changing conditions

Noisy and uncertain observations undermine a robot's state estimation, hindering its ability to perform manipulation tasks correctly

Our research mission

is to expand scientific and technical knowledge in the field of underwater manipulation,

more generally, in marine robotics, with an eye to pursuing new solutions which are "disruptive" at the scientific and technical level.

our development platform

ROV/AUV Hibrid

All techniques and algorithms will be adopted and tested on the e-URoPe platform, an ROV/AUV hybrid




giorni di sviluppo
ricercatori coinvolti
Our approach


The prevailing mechatronic design paradigmis affected by serious issues, even when it is integrated and augmented by Machine Learning and Probabilistic Robotics technologies. Those doubts have surfaced also on general public media. Within the mechatronic paradigm approach if you want to implement more complex behaviors you will have the complexity of your system growing more quickly.

Fabio Bonsignorio

Conventional robots are characterized by rigid links, joints and end-effectors, i.e. manipulators and other tools for interacting with objects. Compliance with the environment is guaranteed through high performance tactile sensors and control systems.
The introduction of mechanically soft links, joints, actuators and end-effectors will decrease the complexity and cost of the system. Moreover system mechanical robustness will increase.

Massimo Caccia

Non-conventional algorithms are cutting-edge solutions to be explored in order to make mobile robots effectively deal with real world situations, given the high degree of uncertainty and unpredictability. In real applications, in which sensor measurements are affected by significant errors, such non-conventional techniques allow to plan the desired reaching movement. The approach can be easily extended to lighter soft robotics arms, characterized by uncertain kinematics and dynamics.

Enrica Zereik

I see a future in which one person will think of an object and he/she will have it in its hands in a few minutes. I see a future where, exploiting the 3D scanners, an existing object can get a replica in a few minutes. I also see a future where there will be a 3D printer in every house and where everyone, even the less talented, will be given the opportunity to create.

Angelo Odetti

The purpose of scientific communication is to exchange information within a given community, a relatively straight-forward task since both the authors and readers of scientific papers are totally familiar with crunching numbers, graphs, functions and abstracts. It's a different matter altogether when the target to be addressed is the general public. Our prototypes are designed to be easy to assemble, inexpensive and user-friendly. We address a wide range of users who are not necessarily specialized. Making our ideas easy and simple to understand is a definite challenge, which must necessarily be faced if the goal is to convey, disseminate and share the results we have obtained and the discoveries we have made.

Pierpaolo Soria

Iscriviti alla nostra newsletter.

Resta in contatto con noi