A SOLUTION TO THE NONCONVEX LINEARLY CONSTRAINED QUADRATING THRUST ALLOCATION PROBLEM

In oil and gas exploration and exploitation many ships and rigs use dynamic positioning (DP) systems for maintaining a fixed position. The DP system usually contains a thrust allocator, which maps the desired generalized forces into individual actuator commands. For over actuated vessels the solution is usually obtained by some kind of optimization. If the thrust allocation problem constitutes a convex problem, a global solution is found easily. However, this is not always the case in real life, where directional constraints on the azimuthing thrusters imposed by e.g. under water equipment may make the problem nonconvex. This paper presents a mixed integer approach to the nonconvex linearly constrained quadratic thrust allocation problem by switching between convex subproblems. The method is demonstrated by experiments.