Abstract
In a ship DPS equipped with azimuth thrusters, desired forces and moment are realized by means of thrust allocation or control allocation (CA) of determining magnitudes of thrusts and directions of thrusters. The most popular method of thrust allocation is well known as Lagrange multiplier method in the case that thrusters can turn around in all directions to reduce the fuel consumption. Recently thrust allocation problems under constraints on thruster performance are discussed, but almost every solving scheme need heavy computational load for iterative calculations to get an optimal solution. Therefore, a new algorithm solving such a problem in real time is required certainly in the navigation field. In the paper, we propose thrust allocation algorithms that handle the bases in the kernel space of a CA matrix which represents the relationship between the azimuth thrusts and control forces. The proposed algorithms satisfy the azimuth thruster angle constraints and make the real-time calculation possible in a reasonable sampling time. Numerical results are given to show the effectiveness of the proposed method, compared with the conventional Lagrange multiplier method.
