Laura Antanas
Papers
8
Total Citations
100
H-Index
5
About
Laura Antanas is a robotics and machine learning researcher whose work sits at the intersection of probabilistic reasoning, relational learning, and robot manipulation. Her research is primarily focused on enabling robots to perform intelligent, context-aware grasping by integrating high-level semantic reasoning with low-level perceptual data — a challenge that demands sophisticated approaches bridging logic, geometry, and learning. Her most influential contribution, "Semantic and Geometric Reasoning for Robotic Grasping: A Probabilistic Logic Approach" (2018, 45 citations), exemplifies her signature methodology: combining probabilistic logic with spatial and semantic information to guide robotic decision-making. This work, along with her earlier exploration of graph kernels for object category prediction (2013, 19 citations), established her as a notable voice in task-dependent grasping research, where simply achieving a stable grasp is insufficient without understanding object properties and task constraints. Antanas has also made meaningful contributions to hierarchical image understanding through relational and distance-based frameworks, advancing scene comprehension for robot vision applications. Her consistent use of Statistical Relational Learning (SRL) across domains — from grasping pipelines to door-opening tasks — reflects a coherent research vision: equipping robots with structured, interpretable, and generalizable world models that make autonomous manipulation both practical and principled.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6Relational Kernel-Based Grasping with Numerical Features5 citations · 2016
- 7Opening Doors: An Initial SRL Approach4 citations · 2013
- 8Relational Affordance Learning for Task-Dependent Robot Grasping4 citations · 2018