Max Robotics — Global Service Robot Directory + US Certification

About

Antoine Laurens is a leading roboticist at the forefront of integrating foundation models with physical manipulation. His research bridges adaptive motion planning, reinforcement learning, and large multimodal models to create robots that can learn and generalize across diverse tasks and embodiments. Laurens’s seminal work on *From Human Physical Interaction to Online Motion Adaptation* (26 citations) introduced parameterized dynamical systems for real-time human-robot collaboration, enabling robots to fluidly adjust tasks during physical interaction. He later tackled the complex challenge of *Robotic Stacking of Diverse Shapes* (16 citations), demonstrating a vision-based RL approach that moves beyond simple pick-and-place. As a core contributor to the landmark *RoboCat* project (9 citations), he helped pioneer a self-improving generalist agent capable of mastering novel skills across multiple robot platforms. Most recently, his work on *Gemini Robotics* (2025) and *DemoStart* (2025) pushes the boundaries of sim-to-real transfer and dexterous manipulation, using demonstration-led auto-curricula to train multi-fingered hands. Laurens’s trajectory—from adaptive control to generalist AI agents—positions him as a key architect of the next generation of intelligent, adaptable robots.