SPECIAL SESSION #05

Angelo Cardellicchio

Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing - National Research Council of Italy (CNR-STIIMA)

• angelo.cardellicchio@cnr.it

Vito Renò

Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing - National Research Council of Italy (CNR-STIIMA)

• vito.reno@cnr.it

Annalisa Milella

Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing - National Research Council of Italy (CNR-STIIMA)

• annalisa.milella@cnr.it

The rapid adoption of Agriculture 4.0 has led to the increasing deployment of autonomous robots powered by complex deep learning models directly in the field. While these robotic systems have demonstrated success in tasks such as harvesting, weeding, and phenotyping, their decision-making processes remain opaque. Thus, the black-box nature of such processes poses significant challenges regarding safety, trust, and reliability, which are further exacerbated by the presence of humans working alongside these robots. Therefore, the research focus has shifted towards the following critical question: why did the system make that decision? To provide an overview of the current advances, this Special Session focuses on the intersection of Explainable AI (XAI) and Agricultural Robotics, aiming to bridge the gap between robotic and algorithmic performance, as well as the interpretability and trustworthiness of the decision-making process. This includes the latest advancements in visualizing how robots perceive their environment, explainable path planning, and transparency in manipulation and decision-making. By making autonomous behaviors trustworthy, the reliability of robotic actions can be validated, thus ensuring safer collaboration and fostering the acceptance of autonomous systems by farmers, breeders, and agronomists.

Topics of interest for this Special Session include, but are not limited to:

• Explainability in Artificial and Robotic Vision for Agriculture;
• Interpretable In-field Navigation;
• Trust in Collaborative Agriculture Robotics;
• Explainable Open Field Harvesting and Weeding;
• Edge-XAI for Agricultural Robotics;
• Phenotyping Reliability;
• Ethical Robotics for Agriculture;
• Uncertainty Quantification in AI-Drive Sensing for Agriculture;
• Interpretable Multi-Modal Sensor Fusion;
• Enhanced Trustworthiness via eXtended Reality.

Dr. Angelo Cardellicchio was born in Taranto, Italy, in 1985. He received a Master's Degree in Computer Engineering from Politecnico di Bari and then a Ph.D. degree in Electrical and Information Engineering from the same University in 2019, defending the thesis "Smart sensor systems for environmental monitoring: implications and Applications." His professional experiences range from web and mobile development to data analysis in the energy, environmental, vulnerability, and industrial domains. He has also been a contract professor for the University of Bari and the University of Foggia, and he is currently a contract professor for the Politecnico of Bari. He has worked for the National Research Council of Italy at the Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing since 2021. He has authored several contributions to international peer-reviewed journals, conferences, and book chapters.

Dr. Vito Renò was born in Bari, Italy, in 1988. He received the Master Degree in Computer Engineering with honors from Politecnico di Bari in 2011, defending a thesis about computer vision and robust background modeling. He also received his PhD degree in Electrical and Information Engineering from the same University in 2017, defending the thesis "3D modeling, reconstruction and analysis of environments assisted by multi-sensorial data processing". He currently is a researcher at CNR STIIMA and is involved in research activities in the fields of computer vision and pattern recognition. He is co-author of 70+ scientific papers and one international patent. Deeply curious and enthusiast about artificial intelligence, with a pinch of multi-disciplinary and synergic applications.

Dr. Annalisa Milella is a Senior Researcher with the Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), National Research Council (CNR) of Italy. She received the Laurea (summa cum laude) and the Research Doctorate degrees in mechanical engineering from the Politecnico of Bari, Italy, in 2002 and 2006, respectively. From 2006 to 2009, she was a Postdoctoral Researcher with the CNR Institute of Intelligent Systems for Automation (ISSIA). From December 2009 to June 2018, she was a permanent Researcher with ISSIA. Since June 2018, she has been a Researcher with CNR-STIIMA. She was the Principal Investigator of the CNR research unit for the H2020 project ATLAS and the ERA-NET ICT-AGRI projects S3-CAV and ANTONIO. Currently, she is the PI of the CNR research unit for the Horizon Europe funded project AgRibot. She is the author of more than 90 publications in international journals, conference proceedings, and book chapters. Her main research interests include multi-sensor systems for robot perception in unstructured environments, 3D reconstruction and mapping, signal and image processing applied to robotics and intelligent systems, and agricultural robotics.