Trajectory of Humanoid Robots with the Unscented Kalman Filter

摘要

In this paper, the integration of the Unscented Kalman Filter (UKF) with gait control for humanoid robot navigation was explored, specifically for the NAO robot. The UKF improves the accuracy of state estimation by processing nonlinear transformations, while gait control adjusts the robot's trajectory using real-time feedback from landmarks detected through computer vision. Experimental results demonstrate that this approach significantly reduces trajectory deviations, leading to improved navigation accuracy and system robustness. This method has potential applications in autonomous robotic systems.