Luca Castri
Papers
6
Total Citations
25
H-Index
4
About
Luca Castri is a researcher at the intersection of robotics, causal inference, and human-robot interaction. His work focuses on developing intelligent systems that can understand and predict human behavior in shared environments, with particular emphasis on causal discovery from time-series data and neuro-symbolic reasoning. Castri’s major contributions include the creation of CAnDOIT, a framework for causal discovery that integrates both observational and interventional data from time series, addressing the critical challenge of hidden confounding factors. His research on causal discovery of dynamic models for predicting human spatial interactions has garnered 8 citations, demonstrating its foundational impact. Castri also pioneered a neuro-symbolic approach to enhance human motion prediction, combining neural networks with symbolic reasoning to improve contextual understanding. His experimental validation of ROS-Causal in real-world human-robot scenarios provides practical insights for deploying robots in crowded environments. With recent work on hierarchical risk-aware navigation in complex intralogistic settings, Castri continues to advance the safe and predictive capabilities of autonomous systems operating alongside humans.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3A Neuro-Symbolic Approach for Enhanced Human Motion Prediction5 citations · 2023
- 4
- 5Qualitative Prediction of Multi-Agent Spatial Interactions2 citations · 2023
- 6