Max Robotics — Global Service Robot Directory + US Certification

About

Robert Platt is a leading researcher in robotic manipulation, with a focus on leveraging geometric symmetries and tactile sensing to make robots more dexterous and sample-efficient. His work centers on equivariant neural networks—models that inherently respect rotational and translational symmetries—which dramatically improve data efficiency for grasp detection, policy learning, and imitation learning. Platt’s contributions include pioneering methods for tactile pose estimation and force-aware policy learning, enabling robots to manipulate unknown objects in unstructured environments. His papers on equivariant models for pick-and-place and reinforcement learning have garnered significant attention, with his most-cited work, “Tactile Pose Estimation and Policy Learning for Unknown Object Manipulation” (11 citations), exemplifying his impact. He has also developed practical assistive technologies, such as a scooter-mounted robot arm for activities of daily living. Platt’s research consistently pushes the boundaries of sample efficiency, from one-shot imitation learning to simulation-augmented training, making his work essential reading for anyone interested in building robust, real-world robotic systems.