Implementation of a cubature points transform.
More...
#include <CubaturePoints.hpp>
|
| template<typename... Dist, std::enable_if_t<(gaussian_distribution< Dist > and ...) and(sizeof...(Dist) > 0> |
| static auto | sample_points (const Dist &...ds) |
| | Calculate the scaled sample points, given a prior distribution and noise terms. More...
|
| |
| template<std::size_t dim, typename YMeans > |
| static auto | weighted_means (YMeans &&y_means) |
| | Calculate the weighted average of posterior means for each sample point. More...
|
| |
|
template<std::size_t dim, typename InputDist , bool return_cross = false, typename X , typename Y , std::enable_if_t< typed_matrix< X > and typed_matrix< Y > and(index_dimension_of< X, 1 >::value==index_dimension_of< Y, 1 >::value) and(index_dimension_of< X, 1 >::value==dim *2) and compares_with< vector_space_descriptor_of_t< X, 0 >, typename DistributionTraits< InputDist >::StaticDescriptor >, int > = 0> |
| static auto | covariance (const X &x_deviations, const Y &y_deviations) |
| |
Implementation of a cubature points transform.
This is as implemented in I. Arasaratnam & S. Haykin, Cubature Kalman Filters, IEEE Transactions on Automatic Control, vol. 54, pp. 1254-1269, 2009.
- Note
- This class has only static members, and is not to be instantiated.
◆ sample_points()
template<typename... Dist, std::enable_if_t<(gaussian_distribution< Dist > and ...) and(sizeof...(Dist) > 0>
| static auto OpenKalman::CubaturePoints::sample_points |
( |
const Dist &... |
ds | ) |
|
|
inlinestatic |
Calculate the scaled sample points, given a prior distribution and noise terms.
The mean of the sample points is effectively translated the origin.
- Template Parameters
-
| Dist | The prior distribution and any optional noise distributions. |
- Returns
- A tuple of sample point matrices, one matrix for each input and noise distribution. Each column of these matrices corresponds to a sample point.
◆ weighted_means()
template<std::size_t dim, typename YMeans >
| static auto OpenKalman::CubaturePoints::weighted_means |
( |
YMeans && |
y_means | ) |
|
|
inlinestatic |
Calculate the weighted average of posterior means for each sample point.
- Template Parameters
-
| dim | The total number of dimensions of all inputs. |
| Arg | A matrix in which each column corresponds to a mean for each sample point. |
- Parameters
-
- Returns
The documentation for this struct was generated from the following file:
