Last November 28th, our head of unit Andrea Micheli was invited to present at the 23rd International Conference of the Italian Association for Artificial Intelligence, in Bolzano. In the Talk Against the clock: lessons learned by applying temporal planning in practice, he focused on his research on the application of temporal planning to real-world scenarios, highlighting the open research directions in this field. Starting from a series of projects carried out by our unit in the past years in different application domains —including robotics, manufacturing, and logistics— Andrea Micheli explored key challenges encountered, the (sometimes hard) lessons learned, and the techniques, tools, and methodologies that have emerged from these efforts. Finally, he also introduced and discussed preliminary results on applying Reinforcement Learning techniques to tailor temporal planners to specific application contexts, which is the main research theme of the ERC project started at the beginning of 2024.
During the same conference, more specifically within the 12th Italian Workshop on Planning and Scheduling, our PhD students, Irene Brugnara and Alessandro La Farciola had the opportunity to present the preliminary results of their research. In particular, they presented the works Synthesis of Domain-Specific Temporal Planning Guidance using Reinforcement Learning: What to Learn and How to Plan and Automatic Selection of Macro-Events for Heuristic-Search Temporal Planning, respectively, both co-authored with Alessandro Valentini and Andrea Micheli.