Jianlang Hu

About

Jianlang Hu is a researcher specializing in robotics, autonomous navigation, and sensor fusion, with a particular focus on visual-inertial odometry (VIO) and state estimation. His work bridges classical estimation theory and modern learning-based approaches to improve robot localization in challenging environments. Hu’s most-cited paper, "Right Invariant SE₂(3)-EKF for Relative Navigation in Learning-based Visual Inertial Odometry" (2022), introduces a novel framework that combines the robustness of learning-based VIO—which excels under varying lighting conditions and without sensor calibration—with the mathematical rigor of invariant extended Kalman filtering on the SE₂(3) Lie group. This contribution addresses a critical gap: while learning-based methods avoid manual calibration and adapt to diverse conditions, they often lack the geometric consistency needed for reliable long-term navigation. By integrating right-invariant error dynamics, Hu’s work enhances relative pose estimation accuracy and consistency, offering a principled solution for autonomous robots. Though early in his career, with 4 citations to date, this work signals a promising trajectory in advancing robust, learning-driven state estimation for real-world autonomy.

Research Focus

Key Achievements

Top Papers

1. 1
   Right Invariant SE₂ (3) - EKF for Relative Navigation in Learning-based Visual Inertial Odometry

   4 citations · 2022

Key Collaborators

Contact & Links