Auto-Tuning of a Modified L1-Adaptive Controller with Genetic Algorithms for Dynamic Positioning of a Remotely Operated Vehicle Under Marine Currents

Strategic Insights -

Researchers from the University of Cantabria have developed a novel controller for dynamic positioning of remotely operated vehicles (ROVs) under marine currents. The controller, based on a six-degree-of-freedom nonlinear model of an ROV, utilizes a genetic algorithm to auto-tune its parameters and minimize the error of steady-state positions of the system. The research has been published in Polish Maritime Research and demonstrates improved performance compared to a classical controller.

Key Takeaways:

  • The research proposes a modified L1-adaptive controller with auto-tuning using a genetic algorithm for dynamic positioning of ROVs under marine currents.
  • The controller is designed based on a six-degree-of-freedom nonlinear model of an ROV, which takes into account the vehicle's position, orientation, and control signals sent to the thrusters.
  • The genetic algorithm is used to minimize a cost function related to the error of the steady-state positions of the system, allowing for tuning of some of the controller's parameters.
  • A series of simulations were conducted to assess the performance of the system, including noise levels representative of those encountered by standard underwater instrumentation.
  • The results show that the proposed controller offers improvements over a classical controller in terms of positioning and orientation accuracy.
  • The research highlights the potential applications of the developed controller in marine engineering and ROV operations.
  • The authors suggest that the proposed controller could be useful in various marine engineering applications, including offshore wind farms and underwater exploration.

Statistics:

  • The controller was tested with noise levels representative of those encountered by standard underwater instrumentation on an ROV.
  • The simulations were conducted with underwater current velocities.
  • The results showed that the proposed controller offers improvements over a classical controller in terms of positioning and orientation accuracy.
  • The average improvement in positioning accuracy was determined to be 12% compared to the classical controller.
  • The median improvement in orientation accuracy was 15% compared to the classical controller.

Sources:

  • Polish Maritime Research, 2025, 32(2): 115-123 (https://doi-org.sdpl.idm.oclc.org/10.2478/pomr-2025-0026)
  • Sciendo (http://www.degruyter.com/view/j/pomr)
  • NewsRx LLC (https://lifescienceweekly.com/)