Outlier-Robust Filtering Based on Robust Similarity Measure
组织者
演讲者
陶飏天择
时间
2023年01月31日 21:00 至 21:30
地点
Online
摘要
In this presentation, we develop a novel outlier-robust Kalman-type filtering framework based on robust similarity measure induced nonlinear regression. This framework reveals the interesting relations existing in current popular robust Kalman-type filters, i.e, Huber-based filter, maximum correntropy based filter.