Max Robotics — Global Service Robot Directory + US Certification

About

Weixiao Liu is a robotics researcher whose work bridges computer vision, manipulation, and human-robot interaction. His key research areas include robotic perception, learning from demonstration, and autonomous surgical systems. Liu’s major contributions span several impactful directions. He developed augmented reality-assisted visual servoing for 6-DOF robotic stereo flexible endoscopes, enabling quicker view adjustment during surgery (26 citations). He introduced Marching-Primitives, a method for shape abstraction from signed distance functions that represents complex objects with compact geometric primitives for efficient physics simulation and robotic manipulation (11 citations). His PRIMP framework uses probabilistically-informed motion primitives for efficient affordance learning from demonstration, learning trajectory distributions in 6D workspace (10 citations). Liu also proposed a learning-free grasping method using hidden superquadrics for unknown objects, eliminating the need for complete 3D models (6 citations). His work consistently addresses practical challenges in robotics—from surgical instrument tracking to object manipulation—by combining geometric reasoning with probabilistic methods. Liu’s research demonstrates how principled mathematical approaches can make robotic systems more autonomous, adaptive, and deployable in real-world settings.