Xinjilefu

Carnegie Mellon University, McGill University

Papers

Total Citations

H-Index

About

Xinjilefu is a leading figure in humanoid robotics, renowned for advancing state estimation and control for dynamic locomotion. His primary research focuses on developing robust algorithms that enable bipedal robots to walk, balance, and recover from disturbances with human-like agility. His most influential work, "State estimation of a walking humanoid robot" (2012, 37 citations), provides a foundational comparison of Kalman Filter designs for walking systems. By contrasting a simple Linear Inverted Pendulum Model (LIPM) with a more complex planar dynamics approach, he demonstrated that the LIPM-based filter offers superior robustness to modeling errors—a critical insight for real-world deployment. This work has shaped how researchers approach sensor fusion and state estimation for legged robots. Additionally, his earlier study on "Hybrid Stabilizing Control for the Spatial Double Inverted Pendulum" (2010) explores control strategies for underactuated systems, further contributing to the theory of balancing and stabilization. Through his rigorous, practice-oriented research, Xinjilefu has helped bridge the gap between theoretical control and reliable robot hardware, making him a key contributor to the field of dynamic walking.