At PAL Robotics, we are partners in a number of EU projects, which are very important in developing robotics for diverse scenarios and new use cases. This blog is a spotlight on TRAIL, a brand-new EU project that is aiming to build AI transparency in robotics.

More on EU project TRAIL focusing on transparency in deep learning, AI, and robotics systems

EU Project TRAIL is an interdisciplinary initiative focusing on AI transparency in robotics, involving experts in AI and robotics who train Doctoral Candidates in designing transparent neural systems. This project not only explores AI decision-making processes but also prepares researchers for impactful roles in the field.

PAL Robotics’ Role in the TRAIL project: developing an Ethical Black Box for service robots

In TRAIL, PAL Robotics plays an important role by developing an Ethical Black Box, advancing AI transparency in robotics, especially in service robots. This initiative combines expertise and mentorship, as we collaborate with specialists and guide our new Ph.D. student. Our objective is to make robot decision-making processes transparent, especially in systems using complex AI. This Black Box will be integrated into our robots like TIAGo, ARI, and StockBot, aiming to enhance their transparency and ethical standards.

Interview with Ferran Gebelli, Ph.D. Candidate and member of the TRAIL project

We took the opportunity to have a chat with our new colleague and PhD Candidate, Ferran Gebelli, about his academic background, as well as his aims for the TRAIL project.

The image depicts Ferran Gebelli interacting with a TIAGo and ARI robots by PAL Robotics. The setting appears to be a busy office or research lab with several individuals at their workstations in the background. The focus is on Ferran and the robot, highlighting a moment of human-robot interaction.

Can you share your academic background and expertise?

“I studied Industrial Engineering at Universitat Politècnica de Catalunya, with a focus on robotics, during my Master’s. I also completed a second Master’s degree in Artificial Intelligence at KU Leuven University. My previous experience includes work in the automotive industry, particularly with autonomous vehicle functionalities, and with Autonomous Mobile Robots in a research center in Belgium.”

What is the importance of the TRAIL project from your perspective?

“AI is expanding rapidly, but often lacks transparency in its models. In robotics, understanding the decisions and behaviour of robots is crucial. TRAIL addresses this need by enhancing transparency in AI and robotics.”

How is AI transparency being addressed in your work within the TRAIL project?

“My focus is on enhancing the transparency of Human-Robot Interaction. It’s about making robots explain their actions in a way that is understandable to different users. The challenge is to provide just enough information for understanding, without being overwhelming or raising any privacy concerns.”

What do you hope to achieve with the TRAIL project?

“I aim to contribute to developing tools and methods that make it easier for people to use and understand robots. This includes making AI in robots more transparent and adaptable for various robotic applications.”

Are you excited to learn from the project supervisors you’ll work with at PAL Robotics, and what do you hope to learn from them?

“Absolutely. The senior research scientists here, with their extensive experience in HRI (Human-Robot Interaction), offer invaluable guidance. I’m eager to learn not only the technical aspects of working with robots and conducting studies, but also to develop a broad range of soft skills that are crucial in a Ph.D. program.”

The TRAIL project, with its focus on transparency in AI and robotics, is a big step forward in our continuous efforts to bring more intuitive and trustworthy robots into everyday life. At PAL Robotics, we remain dedicated to pushing the boundaries of what is possible, always with an eye towards the ethical implications and practical applications of our work. Stay tuned for more updates and insights as we progress further with the TRAIL project.

At PAL Robotics we are partners in collaborative projects in healthcare and Ambient Assisted Living, as well as areas such as agri-food, smart cities, factories of the future, Artificial Intelligence, and Industry 4.0 and 5.0. For more information about PAL Robotics and our involvement in other collaborative projects, visit our website and if you have any questions, make sure you get in touch with us.