The correspondence problem is an important problem when the subjects have two different kinematical structures imitating each other. In this new approach, the imitator and the imitate have totally different dynamics systems, the first one is the fluidic system, the second one is the human hand gestures. The motion of the fluidic system is composed of the combination of fluid particles which are used for the discretization of the problem domain. Observed human hand gestures are imitated by these fluid particles by appropriately adjusting the parameters of the Smoothed Particle Hydrodynamics (SPH), which is a particle based Lagrangian method. In this paper we analyzed the dynamics of the SPH, and used an artificial neural networks (ANN) based controller which automatically adjust the SPH parameters, specifically the body force, to create the desired hand preshapes.