Table 4 Comparative analysis of multimodal trajectory prediction models.
From: Prediction of surface drifter trajectories in the South China sea using deep learning
6 h | 12 h | 18 h | 24 h | |||||
|---|---|---|---|---|---|---|---|---|
RMSE | MAE | RMSE | MAE | RMSE | MAE | RMSE | MAE | |
RNN_Sea | 0.0590 | 4.8727 | 0.1075 | 9.007 | 0.1614 | 13.4843 | 0.2315 | 17.5756 |
LSTM_Sea | 0.0506 | 4.2196 | 0.0901 | 7.4240 | 0.1369 | 11.0177 | 0.1803 | 13.9922 |
GRU_Sea | 0.0744 | 6.1434 | 0.1277 | 10.8076 | 0.2169 | 17.5058 | 0.2689 | 21.2669 |
Transformer_Sea | 0.0630 | 6.0744 | 0.0681 | 6.4069 | 0.1023 | 9.4577 | 0.1791 | 17.8375 |
hycom | 0.1145 | 8.0369 | 0.2166 | 14.5294 | 0.3098 | 20.4355 | 0.3961 | 26.4530 |
hycom_mean | 0.1101 | 7.7895 | 0.2077 | 14.3548 | 0.2953 | 20.2467 | 0.3763 | 25.9198 |
Informer-CNN | 0.0468 | 4.0633 | 0.0575 | 4.8155 | 0.1012 | 8.4299 | 0.1645 | 12.9916 |
