An Evolutionary Solution for Forecasting the Movement of a Moving Object

Abstract

An evolutionary solution for forecasting the movement of a moving object is proposed in this paper. Moving object prediction addresses the problem of locating a moving object correctly. Therefore, how to effectively forecast or predict the location of moving objects can lead to valuable design for real world applications regarding the issues of moving objects. To address the solution for forecasting problem, we applied evolutionary algorithm (EA) as the target algorithm. The algorithm has powerful techniques to helping stochastic search on the fittest solutions for the problems. A case study is a good means to broaden the understanding of the methods how EA can solve problems. Hence, the objective of this paper is to devise the methods that apply EA to solve the practical problems. Two models associated with the operations of EA are explored to forecast the behavior of a moving object. The prediction of next movement for the agent against a moving object is the first model. The results show significant prediction for the next movement of the agent in ten generations. Due to the time constraint, the second model, which concerns with multi-object parameter of the agent, is only in paper specification. In both models, the operations of EAs perform the key roles in searching the fittest forecasting results. The advantages and characteristics of applying EAs on practical problems are observed through the structure design of the models.