CARAML Project

The CARAML project explores new Artificial Intelligence approaches at the intersection of Machine Learning and Symbolic Reasoning. Our research focuses on neuro-symbolic AI — combining neural models with high-level abstractions and formal reasoning — to improve robotic perception and decision-making in complex, real-world environments.

CARAML aims to develop novel AI methods for robotic grasping and manipulation of everyday objects under uncertainty, addressing both the geometric and semantic understanding of previously unseen objects. The project also studies explainable and interpretable reasoning processes, bridging the gap between perception, cognition, and action.

The CARAML team consists of senior and junior researchers, software engineers and external advisors working across fields such as robotics, commonsense reasoning, and machine learning.

• Read More
• Meet the Team