Max Robotics — Global Service Robot Directory + US Certification

About

Ferran Alet is a researcher at the forefront of robotics and machine learning, whose work bridges the gap between perception, manipulation, and meta-learning. His most impactful contributions center on enabling robots to grasp and recognize novel objects in cluttered environments without task-specific training data. His seminal paper on "Robotic Pick-and-Place of Novel Objects in Clutter with Multi-Affordance Grasping and Cross-Domain Image Matching" has accumulated over 700 citations across its versions, establishing a foundational approach for general-purpose robotic manipulation. Alet also made key advances in meta-learning with his work on "Modular meta-learning," which introduced a strategy for learning composable neural network modules that can be flexibly combined to accelerate learning in robotics and other structured prediction domains. Additionally, his exploration of graph neural networks for spatial modeling in "Graph Element Networks" offers a novel framework for adaptive computation without a priori graphical structure. Alet’s research is distinguished by its practical focus on real-world robotic systems and its theoretical depth in learning algorithms, making him a notable figure in the integration of AI and robotics.