AI-PRISM

#AIPRISM25 : Technology provider, research & development – Fondazione Bruno Kessler

The AI-PRISM consortium has the expertise to improve the EU Smart manufacturing value chain in four domains: collaborative environment, robot perception and recognition, robot programming, and human factors. We aim to balance the complementary and overlapping efforts of our industrial partners.

In the project’s most technical and research part to develop our human-centered collaborative robotic platform and AI-enhancing tools, we count on experienced entities such as the Italian partner Fondazione Bruno Kessler (FBK). Thanks to this top Research Institute in Italy ranked 1st place for scientific excellence within three different subject areas and for the economic and social impact according to the latest quality of research ANVUR evaluation, we will excel in our innovative systems and infrastructures.

With its 3,500 square meters of laboratories and scientific infrastructures and a community of over 400 researchers, 140 doctoral students, 200 visiting fellows and thesis students, 700 affiliates and accredited students combined, our partner FBK acts as a scientific and technological hub. They have 11 centers dedicated to technology and innovation and the humanities and social sciences.

At AI-PRISM, we are working with the Digital Industry Center, whose research focuses on digital technologies for the various domains in the industry, including manufacturing, by creating applications for critical systems, adaptive and autonomous systems, advanced perception, diagnostics quality control, and prediction systems. Therefore, they are ideal partners for our dedicated task on ambient digitalization for human-robot collaboration, where automated sensor health and (re)calibration algorithms will be implemented to create a rich network of interrelated signals in the same reference system.

Today the FBK team is sharing a video showcasing their incredible more than 30 years of experience in AI research in Italy, Europe, and the world. This expertise is needed at AI-PRISM to achieve our goal of developing an ecosystem of human-robot collaboration where speed and versatility are required to perform tasks.

In addition to working on the platform, FBK will contribute to our AI-enhancing tools by equipping the robotic agents with perception and planning capabilities. Moreover, FBK will be strongly involved in the development of the AI-PRISM reasoning modules, letting such robotic agents collaborate with human operators to complete assigned tasks. Developing AI-based modules that aim to take the implemented sensor inputs will turn them into high-level descriptors revealing the exact dynamics of what people are doing during their working tasks. Thanks to these fusion modules that integrate raw data to create complex point-cloud data structures, we can represent different perception abilities that, when composed, can describe human activities. This description is then combined with the information of the operating environment to perform both an agent and an ambient level reasoning, acting, and control.

To overcome technical barriers, at AI-PRISM, all implemented sensors will be easy to deploy, configure and integrate with existing factory infrastructures. As a result, FBK will develop APIs to enable the hot swapping of sensors and their auto-recalibration. All the raw and derivative data will be made available to other processing modules, such as the AI-based perception modules or AI-based Controllers, to create new functionalities such as human activity detection or human-aware control solutions.

FBK’s first mission in AI-PRISM is to make our innovations closer and safer for workers. By understanding human activities, robots can mimic and safely collaborate with humans in the workplace. To complement the activity understanding modules, FBK will also develop a health-status component capable of estimating the physical and mental stress level that the worker is facing. A further part of the perception process will also involve modeling worker behavior and health state to predict their future status.

The second mission is to leverage this information to offer a holistic planning module capable of solving multi-agent task planning problems while optimizing available resources, minimizing process duration, and, as a result, improving the working lives of human operators within the factory.

 

Do you want to learn more about AI-PRISM and the team behind the project? Subscribe to our newsletter and follow us on LinkedIn and Twitter, so you don’t miss a thing!

Privacy Preference Center