Research Article
Robust Unscented Kalman Filter for Target Tracking Based on Mahalanobis Distance
@INPROCEEDINGS{10.1007/978-3-030-94182-6_33, author={Bingbing Gao and Wenmin Li and Longqiang Ni and Wei Wang}, title={Robust Unscented Kalman Filter for Target Tracking Based on Mahalanobis Distance}, proceedings={IoT and Big Data Technologies for Health Care. Second EAI International Conference, IoTCare 2021, Virtual Event, October 18-19, 2021, Proceedings, Part II}, proceedings_a={IOTCARE PART 2}, year={2022}, month={6}, keywords={Radar target tracking Unscented Kalman filter Abnormal observations identification Mahalanobis distance}, doi={10.1007/978-3-030-94182-6_33} }- Bingbing Gao
Wenmin Li
Longqiang Ni
Wei Wang
Year: 2022
Robust Unscented Kalman Filter for Target Tracking Based on Mahalanobis Distance
IOTCARE PART 2
Springer
DOI: 10.1007/978-3-030-94182-6_33
Abstract
A robust unscented Kalman filtering (UKF) is presented on the basis of the theory of Mahalanobis distance to enhance radar target tracking’s robustness against abnormal observation information. This method firstly uses the concept of Mahalanobis distance to identify the abnormal observation involved in the radar tracking; and then, a scaling factor with robust property is determined and used into the innovation covariance of classical UKF to weaken the negative effect of aberrant observations on system estimation. The designed robust UKF could effectively enhance the filter’s robustness and improve the tracking accuracy of radar system. The simulation outcomes verify that the designed robust UKF has a better performance than the classical UKF and RUKF, leading to superior tracking performance for radar system.
