审查(临床试验)
离群值
托比模型
上下界
卡尔曼滤波器
计算机科学
数学
算法
控制理论(社会学)
协方差矩阵
数学优化
统计
人工智能
数学分析
控制(管理)
作者
Hang Geng,Zidong Wang,Ali Mousavi,Fuad E. Alsaadi,Yuhua Cheng
标识
DOI:10.1109/tsp.2022.3144945
摘要
In this paper, a novel outlier-resistant filtering problem is concerned for a class of networked systems with dead-zone-like censoring under the weighted try-once-discard protocol (WTODP). To describe the phenomenon of dead-zone-like censoring, the sensor output is characterized by the Tobit model in which the censored region is restrained by specified left- and right-censoring thresholds. The WTODP is employed to decide the transmission sequence of sensors so as to alleviate undesirable data collisions. In the case of the measurement outliers, a saturation function is employed in the Tobit Kalman filter structure to constrain the innovations contaminated by the measurement outliers, thereby maintaining satisfactory filtering performance. By resorting to the approach of the matrix inequality, an upper bound is first obtained on the filtering error covariance where the gain matrix of the Tobit Kalman filter is carefully designed to minimize the obtained upper bound. Moreover, the exponential boundedness of the filtering error is analyzed in the mean square sense. Finally, the effectiveness of the proposed outlier-resistant filtering algorithm is verified by three practical examples.
科研通智能强力驱动
Strongly Powered by AbleSci AI