Corrado Pezzato
Papers
10
Total Citations
169
H-Index
7
About
Corrado Pezzato is a robotics researcher whose work sits at the intersection of probabilistic inference, control theory, and autonomous systems. He is best known for advancing **active inference** — a neuroscience-inspired mathematical framework — as a practical tool for robot perception, control, and planning under uncertainty. His highly cited 2021 survey on active inference in robotics and artificial agents (55 citations) helped establish the field's foundations and open challenges for a broader engineering audience. Pezzato has made notable contributions to fault-tolerant control, developing novel active inference formulations that enable robot manipulators to detect and compensate for sensory faults robustly, work that spans multiple publications and reflects a sustained research thread. His 2023 paper integrating active inference with behavior trees (37 citations) demonstrated how complex, reactive robot behavior in dynamic environments can be elegantly framed as free-energy minimization. More recently, he has pushed into sampling-based model predictive control, leveraging GPU-parallelizable physics simulators for real-time robot decision-making, and combining these methods with active inference for reactive task and motion planning. Across his body of work, Pezzato consistently bridges theoretical neuroscience-inspired frameworks with practical robotic applications, making him a distinctive voice in modern autonomous systems research.
Research Focus
Key Achievements
Top Papers
- 1Active Inference in Robotics and Artificial Agents: Survey and Challenges55 citations · 2021
- 2
- 3
- 4
- 5
- 6Multi-Modal MPPI and Active Inference for Reactive Task and Motion Planning10 citations · 2024
- 7
- 8
- 9Unbiased Active Inference for Classical Control4 citations · 2022
- 10