Max Robotics — Global Service Robot Directory + US Certification

About

Ray Zhang’s research lies at the intersection of robot perception, control, and geometric learning, with a particular focus on legged locomotion and surgical robotics. His most influential work advances symmetry-preserving methods for state estimation and control, leveraging the mathematical structure of geometric spaces to improve robustness and accuracy. In his highly cited 2022 article (13 citations), Zhang synthesized progress in symmetry-aware perception and control, establishing a framework that integrates geometric sensor registration with learning-based techniques. His 2021 contributions on legged robot state estimation (10 citations) introduced a deep learning-based contact estimator that bypasses physical sensors, using multi-modal proprioceptive data from joint encoders and IMUs to enable reliable operation in perceptually degraded environments—a critical advance for quadruped robots. Zhang further extended this work with a deep multi-modal contact estimator for invariant observer design (5 citations). More recently, he developed BDIS-SLAM (2024, 8 citations), a lightweight CPU-based dense stereo SLAM system tailored for surgical applications, demonstrating his ability to translate fundamental geometric principles into practical, real-time solutions. Through these contributions, Zhang has established himself as a leading figure in geometry-driven robot autonomy, with his work directly impacting robust state estimation and control in challenging, real-world settings.