Max Robotics — Global Service Robot Directory + US Certification

About

Carl Henrik Ek is a prominent researcher whose work sits at the intersection of robotics, machine learning, and probabilistic inference, with a particular focus on robot grasping, perception, and human-robot interaction. His contributions have fundamentally advanced how robotic systems learn to manipulate objects intelligently, drawing on rich probabilistic and geometric frameworks to bridge the gap between raw sensory data and meaningful action. Ek's most influential work explores grasp generalization across objects, demonstrating how robots can transfer learned grasping strategies to novel items by identifying shared structural parts — an approach that has garnered over 100 citations. His development of metrics for comparing anthropomorphic hand motion capability (92 citations) provided the field with a standardized benchmark that continues to inform prosthetics and robotic hand design. His probabilistic approaches to task-based grasping (78 citations) and Bayesian Network structure learning (57 citations) reflect a sophisticated integration of uncertainty modeling into practical robotic systems. More recently, Ek has advanced 3D shape estimation through sparse Gaussian process implicit surfaces, combining tactile and visual sensing for robust object modeling. Across more than a decade of research, his work has accumulated hundreds of citations, establishing him as a significant voice in intelligent robotics and probabilistic machine learning for physical agents.