Controls

Intro

• Guidance, Navigation, and Control
• Ship Motions

RobotX- Controls

• Underactuated vehicle- aft-facing motor configuration
  • Can move in surge and yaw axes, but not sway axis
  • Controllerss
    • path following controller
    • station-keeping controller
  • line of sight control (LOS) strategy (Fossen underactuated) combined with WAM-V dynamic model (wamv dynamics)

Software (Control) Subsystem

Global Estimation Methodology for Wave Adaptation Modular Vessel Dynamics Using a Genetic Algorithm

Abstract

Determining parameters for a system model for marine vessels becomes more difficult as the model is made more complex. Work has been done to determine the equations of motion, but not to fully define how to estimate all of the system parameters. This work utilizes a global optimization methodology for estimating the system parameters using a genetic algorithm. The optimizer uses training data sets created from a set of ship maneuvering standards to minimize the error in the 3 degree-of-freedom equations of motion. The model has been optimized using a “No Surge-Yaw” model (minimal surge coupling) and a “Full” model (all states have coupling effects to each other) to determine how well each model can be estimated. The “No Surge-Yaw” model had the best results with making a working marine vessel model. The “Full” model was difficult to optimize due to the additional parameters that had unknown, nonlinear constraints. The “No Surge-Yaw” model was compared to linearized, no coupling version of the model that is commonly used. The linearized model vastly overestimated the results in sway and yaw rate motion while the “No Surge-Yaw” captured the expected coupling dynamics that do exist. Overall, the results of this methodology did generate a set of working marine vessel parameters for an unknown, coupled-state dynamic model.

3 DOF Fossen Model: Line-of-Sight Path Following of Underactuated Marine Craft

Abstract:

A 3 degrees of freedom (surge, sway, and yaw) nonlinear controller for path following of marine craft using only two controls is derived using nonlinear control theory. Path following is achieved by a geometric assignment based on a line-of-sight projection algorithm for minimization of the cross-track error to the path. The desired speed along the path can be specified independently. The control laws in surge and yaw are derived using backstepping. This results in a dynamic feedback controller where the dynamics of the uncontrolled sway mode enters the yaw control law. UGAS is proven for the tracking error dynamics in surge and yaw while the controller dynamics is bounded. A case study involving an experiment with a model ship is included to demonstrate the performance of the controller and guidance systems.

Improved Controller

1/24/25 Meeting Notes

• Current GNC is three functions
  • Planning → sets waypoints
  • Guidance → (LoS) sets desired yaw and velocity
  • Control → sets throttles
    • expects yaw and surge velocity
    • should take desired twist as inputs (surge, sway, yaw velocity)
• Velocity controller
  • If needed - wrap with position controller
  • Incorporates hydrodynamic model
    • thrust to kinematic mapping
  • Incorporate propulsion model
    • throttle / velocity —> thrust mapping
• Test suite
• Structural changes