Successful Validation of Data Assimilation Algorithms

Comparison of the velocities computed from a 3D leap frogging vortex ring in the center slice of the volume, measured with Tomographic PTV

Data Assimilation are physics coupled interpolators used to convert PTV data onto a regular grid. Our most advanced approach, Meshless Track Assimilation (MTA), uses the full set of Navier-Stokes-Equations to regularize the Interpolation. In addition to the velocity interpolation, other quantities such as vorticity, pressure or acceleration can be derived directly from the data.

Lagrangian  Averaged Vorticity Deviation of the triple leap frogging vortex ring / side view, each ring is color coded separately after MTA applied.

During the recent Data Assimilation Challenge organized by ONERA, TU Delft and DLR, MTA demonstrated outstanding performance.
More specifically, MTA delivered results for all the different fluid mechanics properties that were very well matched to the ‘ground truth’ solutions. Another point to note was the consistent performance across the wide range of seeding densities that were provided for the challenge participants.

More information:
Volumetric solutions: Tomographic PTV (4D-PTV) – Dantec Dynamics
MTA paper: Meshless track assimilation (MTA) of 3D PTV data – IOPscience
Data Assimilation challenge: 2nd DA Challenge Case 1 TR Results | ONERA flow reconstruction benchmarks