Reduction in the Amount of Data for Data-driven Passivity Estimation

Reduction in the Amount of Data for Data-driven Passivity Estimation
复制标题

DOI:
10.1109/ccta41146.2020.9206302
复制
发表时间:
2020-08
期刊:
2020 IEEE Conference on Control Technology and Applications (CCTA)
影响因子:
--
通讯作者:
Kanami Iijima;M. Tanemura;S. Azuma;Y. Chida
Kanami Iijima;M. Tanemura;S. Azuma;Y. Chida
中科院分区:
其他
文献类型:
--
作者:
Kanami Iijima;M. Tanemura;S. Azuma;Y. Chida

文献摘要

相似文献

讨论了在没有系统模型的情况下,基于输入输出数据的无源性估计。在传统的基于梯度法的数据驱动估计方法中,收敛速度慢,需要进行大量的实验。因此,我们提出了一种改进收敛性的方法,从而减少了估计所需的数据量。
The estimation of passivity based on input–output data without a system model is discussed. In the conventional data-driven estimation method based on the gradient approach, the convergence is slow and a large number of experiments is required. Therefore, we propose a method to improve the convergence, thereby reducing the amount of data required for estimation.