Marine Vessel Thrust Allocation Using Predicted Force Demand
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Decoupled motion control systems for marine vessels have limited capability to handle thruster constraints, leading to mismatches between desired and achieved forces and torques, especially during aggressive maneuvers, while integrated approaches like Model Predictive Control (MPC) are non-modular and difficult to optimize.
Innovation Solution
A motion control system comprising a higher-level motion controller and a lower-level thrust allocation controller, where the higher-level controller determines desired generalized forces and inputs them to the lower-level controller, which generates control signals for individual thrusters, including rotatable or azimuthing thrusters, considering both current and future forces, velocities, and positions to optimize thrust allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a decoupled motion control system is used, then the system is modular and easier to develop, but the capability to handle thruster constraints is limited
Solution Approach 1:
The system is divided into two independent controllers: a high-level motion controller that handles generalized forces and a low-level thrust allocation controller that handles individual thruster distribution. This segmentation maintains modularity while enabling each controller to specialize in its respective function.
Solution Approach 2:
The low-level thrust allocation controller receives predictions of future generalized forces from the high-level controller and uses this advance information to proactively plan thruster allocations that anticipate upcoming constraint situations, rather than reacting after constraints are violated.
2Reliability
If the bandwidth of the high-level controller is limited to avoid thruster orientation issues, then thruster feasibility is maintained, but control performance is degraded
Solution Approach 1:
The low-level controller receives predicted future forces and uses this advance information to proactively adjust thruster allocations before orientation conflicts arise, eliminating the need to limit controller bandwidth while maintaining thruster feasibility.
Solution Approach 2:
The system implements a feedback loop where the low-level controller monitors actual thruster performance and orientation status, and uses this information to continuously adjust allocations, enabling high-bandwidth control while respecting physical constraints.
3Productivity
If a combined MPC-based motion control system is used, then control performance is improved, but the system becomes non-modular and optimization becomes more difficult
Solution Approach 1:
The control system is segmented into two independent controllers with distinct optimization problems: the high-level motion controller optimizes generalized forces for trajectory tracking, while the low-level thrust allocation controller optimizes individual thruster distributions. This segmentation maintains modularity while achieving combined control performance.
Solution Approach 2:
The high-level controller provides predicted future generalized forces to the low-level controller in advance, enabling the low-level controller to perform proactive optimization of thruster allocations without requiring a complex unified optimization framework.
4Ease of manufacture
If traditional thrust allocation is used, then the system is simple to implement, but there is a mismatch between desired and achieved forces during aggressive maneuvers
Solution Approach 1:
The low-level thrust allocation controller receives predictions of future generalized forces and uses this advance information to proactively plan thruster allocations that anticipate upcoming maneuver requirements, eliminating force mismatches before they occur during aggressive maneuvers.
Solution Approach 2:
The system dynamically adjusts thrust allocation based on real-time thruster status and predicted future forces, enabling adaptive optimization that maintains force accuracy during varying maneuver intensities while keeping the implementation relatively simple.
Data Source
AI summary
A high-level motion controller provides a lower-level thrust allocation with generalized forces demanded on the ship for the current instant of time as well as a prediction of future generalized forces {{circumflex over (τ)}D}t=0T demanded over a predicted future time, and optionally one or more other reference predictions, such as a time-varying velocity and position of the ship {{circumflex over (η)}D, {circumflex over (ν)}D}t=0T. When predictions of the future desired forces and torque are made available to the lower-level thrust allocation, the lower-level thrust allocation will be aware of what is needed in the future and use the predictions to make efficient changes towards the desired future references, when it generates control signals to thrusters available in the thruster system to distribute the desired generalized forces into individual forces of the individual thrusters.