GPU-Based Two-Dimensional Flow Simulation Steering using Coherent Structures

We present an exemplary steering system that performs 2D flow simulation and visualization on graphics processing units (GPUs). The topology of a vector field provides the overall structure and therefore lends itself for steering purposes. We build on the concept of Lagrangian coherent structures present as ridges in the finite-time Lyapunov exponent (FTLE). This allows to perform steering with respect to the true time-dependent dynamics in a given time scope. Based on the insights from the FTLE visualization, our CUDA-based implementation allows effective interactive manipulation of boundary conditions such as solid obstacles or velocity profiles.