Contains data structures and algorithms for filtering approaches to
the Simultaneous Localization and Mapping (SLAM) problem (with known data association). This
package contains three types of SLAM filters
for linear-Gaussian environments:
- the {@link javaslam.slam.KalmanSLAMFilter}, which represents its belief state using a full multivariate Gaussian in the moment parameterization;
- the {@link javaslam.slam.InformationSLAMFilter}, which represents its belief state using a full multivariate Gaussian in the canonical parameterization; and
- the {@link javaslam.slam.TJTSLAMFilter} of (Paskin, 2002), which represents its belief state using a thin junction tree.
In addition, it contains two types of filters for nonlinear SLAM environments:
- the {@link javaslam.slam.LinearizedSLAMFilter}, which couples one of the linear-Gaussian filters above with a {@link javaslam.filter.Linearization} technique; and