OpenKalman::UnscentedParametersParameterEstimation Struct Reference

Unscented parameters for use in parameter estimation. More...

#include <Unscented.hpp>

Static Public Attributes
static constexpr double	alpha = 0.001
	Scaling factor (typically 0.001 but may be from 0.0001 to 1).
static constexpr double	beta = 2.0
	Factor to compensate for the distribution (beta==2 is optimal for Gaussian distributions).
template<std::size_t dim>
static constexpr double	kappa = 3. - double(dim)
	Secondary scaling parameter. More...

Detailed Description

Unscented parameters for use in parameter estimation.

Member Data Documentation

kappa

template<std::size_t dim>

constexpr double OpenKalman::UnscentedParametersParameterEstimation::kappa = 3. - double(dim)

static

Secondary scaling parameter.

Usually 0 for state estimation and kurtosis-dim for parameter estimation. This makes it possible to match some of the fourth-order terms. (For Gaussian, kurtosis = 3).

---

The documentation for this struct was generated from the following file:

• OpenKalman/transforms/sample-points/Unscented.hpp