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 2302.02694math.OCcs.SYeess.SPeess.SY
收藏

鲁棒最大相关熵卡尔曼滤波

Robust Maximum Correntropy Kalman Filter

Joydeb Saha, Shovan Bhaumik

摘要:

卡尔曼滤波器为具有高斯噪声的线性系提供最优估计。然而,当噪声是非高斯的性质,其性能迅速恶化。对于非高斯噪声,最大相关熵卡尔曼滤波(MCKF)的发展,提供了一个改进的结果。但当系统模型与标称值不同时,MCKF的性能会下降。对于这种情况下,我们提出了一种新的鲁棒滤波技术,最大化的费用函数定义的指数加权过去和现在的错误以及高斯核函数。通过解决这个成本标准,我们开发了先验和后验均值和协方差阵传播方程。通过最大化误差矩阵的相关熵函数,我们选择了在每个时间步的核带宽值。并进一步给出了算法收敛的条件。给出了两个数值例子来说明新滤波技术的有效性。

英文摘要:

The Kalman filter provides an optimal estimation for a linear system with Gaussian noise. However when the noises are non-Gaussian in nature, its performance deteriorates rapidly. For non-Gaussian noises, maximum correntropy Kalman filter (MCKF) is developed which provides an improved result. But when the system model differs from nominal consideration, the performance of the MCKF degrades. For such cases, we have proposed a new robust filtering technique which maximize a cost function defined by exponential of weighted past and present errors along with the Gaussian kernel function. By solving this cost criteria we have developed prior and posterior mean and covariance matrix propagation equations. By maximizing the correntropy function of error matrix, we have selected the kernel bandwidth value at each time step. Further the conditions for convergence of the proposed algorithm is also derived. Two numerical examples are presented to show the usefulness of the new filtering technique.

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。

↑