Robust EKF EKF-SLAM Method ethod Against gainst Disturbance isturbance Using sing The he Shifted Mean based Covariance Inflation Technique echnique
This paper presents a novel solution to overcome the disturbance noise (outlier) for the Extended Kalman Filter based Simultaneous Localization And Mapping (EKF EKF-SLAM) SLAM). The standard Kalman Filter (KF) is not robust to the he disturbance noise. The possibility that distu disturbance may rbance happen is high high, because SLAM aims at exp exploring loring unknown environment. Hence KF based SLAM methods should consider how to handle the disturbance noise. Variations of KF have been introduced to overcome this problem problem. However, these method . Methods employ manual parameter tuning tuning, detecting/ weighting method method. T . The core of he our algorithm is to inflate the state uncertainty by using the magnitude of innovation innovation, without , tuning and detecting detecting. Although it is impossible to estimate the state value immediately, the in inflated flated state uncertainty makes it possible for the estimat estimated ed value to converge on the true value much faster. We evaluate the proposed method under the well well-known benchmark Matlab program. The result results show that the proposed method overcomes s the disturbance noise and increase increases the performance of s EKF EKF-SLAM SLAM.
Won-Seok Choi Se Se-Young Oh
Department of Electronic and Electrical Engineering,the POhang university of Science and TECHnology (POSTECH),Pohang,KOREA
国际会议
2011 IEEE International Conference on Robotics and Automation(2011年IEEE世界机器人与自动化大会 ICRA 2011)
上海
英文
4054-4059
2011-05-09（万方平台首次上网日期，不代表论文的发表时间）