What is it about?
Showing that a very simple approximate filter can do the job of filtering well enough if we have a low noise chanel. No need for the EKF or other expensive filter
Featured Image
Photo by Adi Goldstein on Unsplash
Why is it important?
It can be used in robotics or any situation where we need to estimate the state (position, velocity) in real time and the available observations are of one quantity only, for instance 1 sensor. We propose an approximate solution and mathematically prove that it is as good as the exact solution. It can naturally be generalized to multiple dimensions (dimension 2n for the state nad dimension n for the observations or sensors)
Perspectives
Explains that good approximations for the estimation/filtering problem exist and there is no need to use computer intensive solution that are unfeasible in real time.
Dr Paula P Milheiro-Oliveira
University of porto
Read the Original
This page is a summary of: Approximate Nonlinear Filtering for a Two-Dimensional Diffusion with One-Dimensional Observations in a Low Noise Channel, SIAM Journal on Control and Optimization, January 2003, Society for Industrial & Applied Mathematics (SIAM),
DOI: 10.1137/s0363012902363920.
You can read the full text:
Contributors
The following have contributed to this page
