Experimental Validation of Wave Induced Disturbances for Predictive Station Keeping of a Remotely Operated Vehicle

Research output: Contribution to journalArticlepeer-review

Abstract

Predictive control methods can substantially improve the performance of Unmanned Underwater Vehicles (UUVs), particularly in shallow water environments or near the free surface where wave induced disturbance are of magnitude comparable to the vehicle characteristic inertia. To facilitate the adoption of these methods, a fast estimation of the time evolution of hydrodynamic forces acting on a vehicle is required. To this end, we perform experiments in a wave tank with an ROV to validate the use of Linear Wave Theory (LWT) to capture the time history of surge, heave and pitch wave induced forces and moments. Validation is performed for various sea states, reconstructed with a mean correlation of 0.9138 in comparison to experimental measurements, displaying a maximum normalised mean error deviation between simulation and experimental data of 0.16 and 0.27 respectively for surge and heave forces, and 0.34 for pitch moment. The effectiveness of employing real-time wave disturbance forecasting for the purpose of anticipatory control is then assessed by incorporating the predicted loads within a Model Predictive Controller. Results display a mean RMS positional error reduction of 47.32% in comparison to a standard PD controller. This presents evidence that accurate, near real-time predictions of the wave-generated forces and moments on an ROV can be produced, laying the foundation for developing model-based predictive control strategies that better suit operation in harsh environments.

Original languageEnglish
Pages (from-to)5421 - 5428
JournalIEEE Robotics and Automation Letters
Volume6
Issue number3
Early online date27 Apr 2021
DOIs
Publication statusE-pub ahead of print - 27 Apr 2021

Keywords

  • Attitude control
  • Disturbance Estimation
  • Estimation
  • Hydrodynamics
  • Marine Robotics
  • Model Predictive Control
  • Motion Control
  • Predictive models
  • Station Keeping
  • Surges
  • Task analysis
  • Vehicle dynamics

Fingerprint

Dive into the research topics of 'Experimental Validation of Wave Induced Disturbances for Predictive Station Keeping of a Remotely Operated Vehicle'. Together they form a unique fingerprint.

Cite this