A mixed finite element discretization of dynamical optimal transport (Q2144978)

The purpose of this paper is to propose a mixed finite element approach to solve the dynamical formulation of optimal transport with quadratic cost on unstructured meshes. The method can be easily modified to include different type of regularizations which are relevant for the dynamic reconstruction and interpolation of physical quantities. The discrete solutions, for specific choices of finite element spaces, is proved to converge towards the solutions of the optimal transport problem between two arbitrary measures. The discrete problem is solved by using a proximal splitting approach. The details on how to introduce regularization terms in the formulation are described. Two numerical tests are presented to demonstrate the behaviour of the proposed discretization both qualitatively and in terms of convergence of the algorithm.