Fig. 7: Varying rotational mobility coefficient. | Communications Physics

Fig. 7: Varying rotational mobility coefficient.

From: Hierarchical equivariant graph neural networks for forecasting collective motion in vortex clusters and microswimmers

Fig. 7

Snapshots of microswimmers at various rotational mobility coefficients (ν). Test data is in black and HE-GNN prediction is in blue. a and b begin from the same random initial condition discussed in the “Microswimmers” subsection of “Methods” section. c Is a time series starting from the end condition of a with a different rotational mobility coefficient.

Back to article page