Unscented Schmidt–Kalman Filter Algorithm

What is it about?

The Unscented Kalman Filter is a non-linear extension of the Kalman filter, which is the standard bearer of filtering. This paper introduces an Unscented Schmidt -Kalman Filter, which allows the filter to "consider" uncertainty in terms that are not being estimated in the filter.

Featured Image

Why is it important?

This algorithm is important because it allows for considering uncertainty in parameters that are not observed adequately enough to estimate. The gravity term or a sensor bias in orbit determination are simple examples. These terms may not be observable by the sensor, which could cause aliasing if one tried to estimate the terms, however, it is known that the values used in the model are not perfect, thus it is desirable to consider these uncertainties.