fdaPDE implements a class of spatial regression models in between statistics and numerical analysis. These models are particularly well-suited for situations where a prior knowledge of the underlying phenomenon is known and can be described in terms of a Partial Differential Equation (PDE).
For the installation details see the file INSTALL.
Some example can be found visualizing the help of the three main smoothing functions i.e.
?smooth.FEM.basis or ?smooth.FEM.PDE.basis or ?smooth.FEM.PDE.SV.basis