filmov
tv
Stable Adaptive Extended Kalman Filter

Показать описание
This video summarizes the topics covered in the paper "A Stable Adaptive Extended Kalman Filter for Estimating Robot Manipulators Link Velocity and Acceleration". This work has been accepted in the IROS 2023 conference.
The link velocity and acceleration of robot manipulators can be accurately and rapidly estimated through the utilization of an extended Kalman filter, offering high bandwidth capabilities. This study enhances the performance of the extended Kalman filter by dynamically adjusting the noise covariance matrices employed within the filter in an online manner. Furthermore, a comprehensive analysis of the filter's stability is presented, along with a novel approach to ensure its stability—a facet that has received limited attention within existing literature. Empirical validation of these propositions is conducted using a 7-degree-of-freedom robot, yielding promising outcomes.
The link velocity and acceleration of robot manipulators can be accurately and rapidly estimated through the utilization of an extended Kalman filter, offering high bandwidth capabilities. This study enhances the performance of the extended Kalman filter by dynamically adjusting the noise covariance matrices employed within the filter in an online manner. Furthermore, a comprehensive analysis of the filter's stability is presented, along with a novel approach to ensure its stability—a facet that has received limited attention within existing literature. Empirical validation of these propositions is conducted using a 7-degree-of-freedom robot, yielding promising outcomes.