Lixin Yang

Papers

Total Citations

H-Index

About

Lixin Yang is a robotics researcher whose work spans robot perception, calibration, and learning-based manipulation. With contributions ranging from foundational sensor-robot integration techniques to cutting-edge generative approaches for robotic control, Yang's research addresses some of the most pressing challenges in enabling robots to understand and interact with their physical environments. Among Yang's notable contributions is a depth camera-based hand-eye calibration framework using a sphere model approach, published in 2018 and accumulating 21 citations. This work tackled the practical challenge of determining the precise transformation between a depth camera and a robot's wrist — a critical prerequisite for accurate robotic manipulation — by leveraging indirect measurements derived from robot motion and geometric calibration tools. More recently, Yang's research has pushed into the frontier of diffusion-based robot learning. The 2025 paper introducing MBA (Motion Before Action) proposes an innovative imitation learning paradigm in which robots first reason about expected object motion from visual observations before generating action sequences. This approach reflects a growing trend toward imbuing robots with richer predictive reasoning capabilities. Together, these works illustrate Yang's consistent focus on bridging perception and action in robotic systems.