3D trajectory optimization of the slender body freely falling through water using cuckoo search algorithm

Abstract The problem of trajectory optimization of underwater slender bodies is commonly encountered during offshore operations such as deploying the anchor into seabed at the minimum locating error and releasing unmanned aerial-aquatic vehicle (UAAV) into water for the largest terminal speed. To solve these problems, an accurate trajectory prediction of the freely falling slender body together with an efficient optimization algorithm are required to build up the optimization framework which aims at satisfying a specific task. In this paper, firstly, the improved 3D dynamic model is proposed which can properly capture the 3D effect by taking into account lateral forces and motions induced by the asymmetric vortex shedding and axial rotations during crossing flow. By simulating the trajectory of a freely falling cylinder, the proposed 3D dynamic model shows much better agreement with the experiment compared with traditional 2D dynamical models which validates its capability into modelling the freely falling motion of the cylinder. The effects of axial rotating rates are systematically studied and discussed for freely falling with different drop angles. In order to obtain the optimal trajectory under a specific optimization index, the Cuckoo Search (CS) optimization algorithm is implemented into the 3D dynamical model by using random walk and Levy flight respectively and compared with two popular algorithms: Genetic Algorithm (GA) and Particle Swarming Optimization (PSO). Because of the stronger exploration ability induced by Levy flight, the proposed CS algorithm with Levy flight is found performing much better than the CS algorithm with random walk and even GA and PSO both in optimizing the unconstrained single objective optimization problem: terminal position error and the constrained single objective optimization problem: the maximum terminal velocity.