Minjie Liu

University of Technology Sydney, The University of Sydney

Papers

Total Citations

H-Index

About

Minjie Liu is a roboticist whose research focuses on advancing Simultaneous Localization and Mapping (SLAM) through innovative environmental representations. Her core contributions lie in developing feature-based SLAM algorithms that maximize information usage from laser sensor data, moving beyond traditional point or line features. In her most-cited work (22 citations), she formulated SLAM as an optimization problem, modeling environments as continuous curves to improve data utilization. She further pioneered the use of B-Splines for statistically consistent SLAM (13 citations), enabling flexible, feature-less environment modeling that avoids geometric extraction errors. Her work on B-Spline SLAM observation models (2009) refined how control points are integrated into the state vector for EKF-based estimation. By replacing discrete landmarks with smooth parametric curves, Liu’s research addresses fundamental challenges in consistency and information retention for laser-based SLAM systems. Her contributions are particularly notable for bridging the gap between raw sensor data and robust mapping, offering a path toward more reliable autonomous navigation in unstructured environments.