No announcement yet.

VRX Simulator Fidelity

  • Filter
  • Time
  • Show
Clear All
new posts

  • VRX Simulator Fidelity

    Hi all. At the recent RobotX Interactive forum held in Singapore, I recall Brian mentioning that a certain type of data regarding thrusters and its performance in the sea waters would help to improve the fidelity of the simulator. However, I can't seem to remember the exact technical term that he had mentioned. Could someone who had attended the forum help me out?

  • #2
    Hi Hashir,

    It was a pleasure to meet you in Singapore. Thank you for bringing this up.

    The general discussion was about characterizing the dynamics of the WAM-V platform. Some of the important model properties are the hydrodynamic coefficients and the thrust curves of the propulsion system. Researchers at FAU have published a few articles that include these details, e.g.,

    title = "Station-keeping control of an unmanned surface vehicle exposed to current and wind disturbances",
    journal = "Ocean Engineering",
    volume = "127",
    pages = "305 - 324",
    year = "2016",
    url = "",
    author = "Edoardo I. Sarda and Huajin Qu and Ivan R. Bertaska and Karl D. von Ellenrieder"

    If anyone in the community has done more model/system identification on the WAM-V platform, having better estimates of the parameters helps us improve the simulation.

    The particular characterization we discussed is the propulsion characterization. A static "bollard pull test" is often done to characterize the propulsion. The Sarda article above reports the results for such a test for the FAU thruster. This is what we used for VRX. I believe FAU has similar results for one of the Torqueedo systes - which will be nice to see. Keep in mind that this simple characterization ignores the influence of speed through the water on the advance ratio and affects do to hull interaction, but it is a reasonable place to start for this application.

    So, if anyone has done any characterization of the WAM-V platform, it would help everyone to have more accurate model coefficients.

    Thank you,