Anchor-Based Adaptation for Modeling Vessel Maneuvering Dynamics

The dataset can be downloaded from: https://github.com/BertVanAcker/rpio-examples/tree/main/NTNU/Realization/ManagedSystem/Simulator/Manuevering_Data_K_SIM

Place the downloaded data inside a data folder following this structure:

data/
├── No_Wind/
├── 2knots/
├── 7knots/
├── 12knots/
└── 17knots/

---

In order to train the anchor models:

  python train_anchor.py --model 'lstm' 
  python train_anchor.py --model 'gru' 
  python train_anchor_eulernn.py 

In order to train the adaptation models:

  python train_adaptation.py --model 'lstm'
  python train_adaptation.py --model 'gru'
  python train_adaptation_eulernn.py
  python train_adaptation_euler_anchor_gru_adapt.py

In order to train the adaptation models:

  python train_residual_refinement.py --model 'lstm'
  python train_residual_refinement.py --model 'gru'
  python train_residual_refinement.py --model 'eulernn' --warmup_period 0
  python train_residual_refinement.py --model 'eulernn-gru'

---

Acknowledgments

The work presented here is supported by the RoboSAPIENS project funded by the European Commission’s Horizon Europe programme under grant agreement number 101133807. This publication reflects the authors’ views only. The European Commission is not responsible for any use that may be made of the information it contains.

Learn more about RoboSAPIENS.