Shaojing Su

National University of Defense Technology

Papers

Total Citations

H-Index

About

Shaojing Su is a leading researcher in robotics and computer vision, with a primary focus on semantic scene understanding and autonomous navigation. Her work bridges the gap between perception and action, addressing critical challenges in how robots interpret and operate within complex, dynamic environments. Su’s major contributions include the development of the Link-RGBD module, a cross-guided feature fusion network that significantly enhances RGBD semantic segmentation by more effectively leveraging depth information to improve scene parsing accuracy. This work has garnered 24 citations, underscoring its impact on the field. She is also the architect of SOLO-SLAM, a parallel semantic SLAM algorithm that overcomes the limitations of traditional systems in dynamic scenes, achieving robust localization and mapping where conventional methods fail. This innovation, with 19 citations, is vital for mobile robots operating in real-world, non-rigid environments. Additionally, Su has advanced nonlinear robot system control with a data-driven Kalman filter using kernel-based Koopman operators, offering theoretical guarantees for systems with unknown dynamics. Her research is instrumental in pushing the boundaries of autonomous robotics, making her a notable figure in the integration of deep learning with classical robotics frameworks.