Max Robotics — Global Service Robot Directory + US Certification

About

Lucas Manuelli is a robotics researcher whose work spans robotic manipulation, object representation, contact estimation, and human-robot interfaces. He is perhaps best known for developing **KPAM (KeyPoint Affordances for Category-Level Robotic Manipulation)**, which redefined how robots specify and achieve manipulation goals by using semantic keypoints rather than rigid 6-DOF pose estimates — a contribution that has garnered over 170 citations and influenced category-level manipulation research broadly. Manuelli's early work introduced the **Contact Particle Filter** (2016, 101 citations), an elegant probabilistic approach to localizing external contacts on robot bodies using only proprioceptive sensing — a foundational capability for safe human-robot interaction. His contributions to **Dense Object Nets** (2018, 100 citations) helped establish dense visual descriptors as a powerful, task-agnostic representation for manipulation, while his **Label Fusion** pipeline (2018, 113 citations) addressed the critical challenge of generating high-quality training data for deep learning in cluttered scene understanding. More recently, Manuelli has explored the intersection of language, vision, and manipulation through works like **CLIPort** (2021, 99 citations) and self-supervised keypoint learning for model-based reinforcement learning. His research consistently bridges perception, representation learning, and practical robot deployment, making him a significant contributor to modern robotic manipulation.