Maria Gustavsson

Rita Cunha, IST, Portugal

Short Bio: Rita Cunha received her Ph.D. degree in Electrical and Computer Engineering from Instituto Superior Técnico (IST), Universidade de Lisboa, Portugal, in 2007. Since 2019 she has been with the Department of Electrical and Computer Engineering of IST, where she teaches in the areas of Control and Robotics. She is currently an Associate Professor of IST, member of the Executive Board of the Department of Electrical and Computer Engineering, and a senior researcher with the Institute for Systems and Robotics (ISR). Her research interests and expertise include dynamical systems, nonlinear control systems, motion planning and control of autonomous vehicles, distributed control systems, vision-based control, and aerial robotics. She has participated in several research projects, including 2 EU projects as coordinator of IST's participation and 4 national projects as principal investigator and 3 national projects as co-PI. Selected projects for which she coordinated IST’s participation include: MULTIDRONE – Multiple Drone platform for media production (H2020), LOTUS - Load Transportation using Unmanned Aerial Vehicles (FCT, Portugal2020), and SCARVE - Sensor-based Control of Autonomous Aerial Vehicles (FCT).

Title: Aerial Robotics: Advances in Motion Planning and Control

Abstract: Unmanned Aerial Vehicles (UAVs), commonly known as drones, are rapidly evolving to become versatile sensing platforms, capable of navigating and tracking trajectories with remarkable accuracy. While motion control in free flight is now well established, new challenges are driving the field toward richer interaction with the environment and greater cooperation between multiple vehicles. In this talk, I will first give a brief overview of past work on trajectory tracking and path following control of quadrotor UAVs, which leverages Lyapunov stability theory to provide high-performance autonomous flight. I will then highlight developments related to three topics of research: i) formation planning and control for multi-vehicle systems; ii) visual servoing control using optical flow for reactive landing, and iii) and motion planning and control for aerial transportation of slung loads.




fillipo sanfilippo

Filippo Sanfilippo, University of Agder, Norway

Short Bio: Filippo Sanfilippo holds a PhD in Engineering Cybernetics from the Norwegian University of Science and Technology (NTNU), Norway, with a focus on intelligent control approaches for robotic manipulators. His research interests include robotics, wearables, human-robot teaming, artificial intelligence, and control theory. He is currently appointed as a Professor at the Faculty of Engineering and Science, University of Agder (UiA), Grimstad, Norway. He is also an adjunct Professor at the Faculty of Informatics, Kaunas University of Technology, Kaunas, Lithuania. He is also the Director of Science at Twilligent AS, Norway, a company that is at the forefront of creating intelligent digital twins to visualise, simulate and optimise the operation of complex facilities, production lines and processes. He carries a vast experience in participating in European research programs and various national projects from the Research Council of Norway (RCN). He is an IEEE Senior Member. He is the former Chair of the IEEE Norway Section. He is the Chair of the IEEE Robotics and Automation, Control Systems and Intelligent Transportation Systems Joint Chapter. He is the Chair of the Norway Section Life Members Affinity Group. He is currently a member of the IEEE Region 8 Chapter Coordination Committee, of the Conference Coordination Committee, of the IEEE Public Visibility Committee, of the IEEE R8 Awards and Recognitions Committee, and of the Professional and Educational Activities Committee. He is also the Treasurer of the Norsk Forening for Kunstig Intelligens (NAIS), the Norwegian Association for Artificial Intelligence. He has authored and co-authored several technical papers in various journals and conferences. He is a reviewer for several international conferences and journals.

Title: Human-Robot Teaming, a Forward Leap into Real Life Applications

Abstract: Human-robot interaction (HRI) is the study of how humans and robots interact, as well as how to develop robots that can adapt to human behavior. Human-robot cooperation (HRC) expands on this by creating new approaches and technologies that allow robots to collaborate with people in shared environments. The field of human-robot teaming (HRT) goes one step further, by studying how to create teams of humans and robots that can work together effectively and efficiently to achieve common goals. In this talk, an overview of the possible real-life applications for HRT will be presented.




José Machado

José Machado, University of Minho, Portugal

Short Bio: José Machado is a Full Professor at the University of Minho, Portugal, and a senior researcher at the ALGORITMI Research Center, where he coordinates the Computer Science and Technology (CST) group and the Knowledge Engineering Laboratory. He is also a member of the coordination committee of LASI-Intelligent Systems Associate Lab. His research spans Artificial Intelligence, Data Science, and Medical Informatics, with a particular focus on intelligent decision-support systems, responsible AI, and digital health. He has led several national and international R&D projects and has supervised numerous Ph.D. and M.Sc. theses. He is actively involved in scientific committees and serves regularly as a reviewer and evaluator for international journals, conferences, and funding agencies.

Title: Low-Power Real-Time Machine Learning Approach using IMU Data on FPGA

Abstract: This talk presents a low-power, real-time machine learning approach for processing inertial data on FPGA platforms. Leveraging data from Inertial Measurement Units (IMUs), the system performs efficient classification tasks directly on hardware, enabling immediate response with minimal energy consumption. Designed for embedded and mobile applications, it addresses critical constraints such as latency, power usage, and data privacy. By executing all processing locally on the FPGA, the solution avoids transmitting raw sensor data to external servers, thereby preserving user privacy. This architecture is particularly suited for smart vehicles, wearable devices, and autonomous systems, where energy efficiency, real-time responsiveness, and privacy protection are essential. Experimental results demonstrate the feasibility of deploying lightweight ML models on FPGA with high performance and privacy-aware operation. .