Max Robotics — Global Service Robot Directory + US Certification

About

Nikolaos Gkanatsios is a rising star in robotic manipulation, focusing on the intersection of 3D perception, language understanding, and action generation. His research addresses two fundamental challenges: enabling robots to understand compositional language instructions and to grasp objects with high spatial precision. In his influential work "Energy-based Models are Zero-Shot Planners for Compositional Scene Rearrangement" (16 citations), he pioneered a framework that generalizes to complex, multi-constraint instructions without task-specific training. His "Act3D" (6 citations) introduced 3D Feature Field Transformers, a computationally efficient architecture for multi-task manipulation that achieves high-resolution spatial reasoning. Gkanatsios has also made key contributions to robotic grasping, developing orientation-attentive methods that resolve ambiguities in grasp synthesis (10 and 4 citations). His most recent work, "3D Diffuser Actor" (4 citations), advances policy diffusion by integrating 3D scene representations, setting a new standard for action distribution learning. Through these innovations, Gkanatsios is shaping how robots perceive, reason, and act in three-dimensional spaces, bridging the gap between language and physical interaction.