arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1401.0077eess.SYcs.SY
收藏

多模态过滤用于非线性估计

Multi-modal filtering for non-linear estimation

Technische Universität Darmstadt(达姆施塔特工业大学);MPI for Intelligent Systems(马克斯·普朗克智能系统研究所)

Sanket Kamthe, Jan Peters, Marc P Deisenroth

AI总结:

本文提出一种非线性滤波算法,利用高斯混合模型表示多模态密度,通过闭式估计参数提升时间序列和实际应用中的跟踪性能。

摘要:

多模态密度在时间序列和实际应用中频繁出现。然而,它们无法由常规状态估计器如扩展卡尔曼滤波器(EKF)和无迹卡尔曼滤波器(UKF)表示,且这些方法还存在不确定性往往无法充分捕捉的问题,可能导致跟踪性能不一致和发散。本文通过设计一种非线性滤波算法,利用高斯混合模型表示密度,其参数通过闭式形式估计。所提出的方法在典型基准测试中表现出优越的性能。

英文摘要:

Multi-modal densities appear frequently in time series and practical applications. However, they cannot be represented by common state estimators, such as the Extended Kalman Filter (EKF) and the Unscented Kalman Filter (UKF), which additionally suffer from the fact that uncertainty is often not captured sufficiently well, which can result in incoherent and divergent tracking performance. In this paper, we address these issues by devising a non-linear filtering algorithm where densities are represented by Gaussian mixture models, whose parameters are estimated in closed form. The resulting method exhibits a superior performance on typical benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑