The Development of Genetic Algorithm for the Prediction of Ship Motion from Wave Radar Data

Genetic Algorithms represent a model of natural process based on the Darwinian principle of evolution. Genetic algorithms are used to solve optimization problems by generating a set of possible solutions and determining which of these solutions most closely match the desired result. The Genetic Algorithm (GA) initially generates a random set of possible solutions, or populations, determines which members of the population are most desirable, and calculates better solutions by implementing genetic operations such as reproduction, cloning, and mutation. Problems where the evaluation of a solution is computationally simple enable the GA to search a vast search space. One of the very important applications of genetic algorithms is in predicting the motion of ship due to the sea conditions or wave structure. This thesis mainly focuses on the development of a GA to predict ship motion from surface wave data. Various GA techniques, their advantages, and drawbacks are discussed. The applicability of these techniques for this application is also investigated. The selection of genetic operators will determine how the characteristics of future members of the population (or children) are related to the suitability of existing (or parent) members. Reproduction is used to obtain a solution that has the direct characteristics from the parents. Crossover is used to get a solution that has a combination of characteristics from both the parents based on certain criteria. Mutation is implemented in situations where diversity is desired to avoid convergence to a local minimum in the search space. In this thesis, an algorithm to predict ship motion based on GA techniques will be presented, and results of the approach for simulated sea states will be discussed. The proposed GA uses three genetic operators (Reproduction, Crossover, and Mutation) that are tested on these sea states with varied configurations of the GA process. This algorithm searches for the best transfer function of the ship motion based on different sea states (ocean surfaces) in order to derive the coefficients for the predicted motion that best represents the actual ship motion. A comparison is then made between the results obtained from different sea states to determine the best GA configuration to provide the most rapid and best fit to the actual ship motion for the selected sea surface. This algorithm has prominent applications in logistics ( cargo transfer between the ships in the middle of the sea) and military purposes (launching ship board missiles). In addition, for high performance vessels, such an algorithm could allow the real time adjustment of heading to avoid severe sea conditions that may damage ship, cargo, or personnel.