Max Robotics — Global Service Robot Directory + US Certification

About

Marco Tamassia’s research bridges reinforcement learning and multi-robot systems, with a focus on enabling autonomous agents to operate effectively in complex, real-world environments. His early work tackles the challenge of visual coverage for search and rescue, where teams of mobile robots must collaboratively capture visual information from large-scale 3D spaces. To address this, he developed novel directional visual descriptors that measure visual information quality, allowing robots to plan efficient coverage paths. This foundational research, published between 2013 and 2015, has practical implications for 3D reconstruction and active surveillance. More recently, Tamassia turned to reinforcement learning, exploring how agents can accelerate learning by acquiring “options”—high-level behaviors—from demonstrations. His 2017 Pac-Man case study (7 citations) shows how imitation learning can bootstrap an RL agent’s performance, reducing the costly trial-and-error phase. Though his citation counts are modest, his work is notable for its clear applied focus: from coordinating camera-equipped robots in disaster zones to teaching game agents smarter strategies. Tamassia’s contributions exemplify how combining geometric reasoning with learning can make autonomous systems more practical and sample-efficient.