Skip to Main Content
Ant colony optimization (ACO) algorithm is usually utilized to solve various combinatorial optimization problems. In this work, however, two novel ant systems are developed to estimate the state of interest, and we call them ant estimators. The first ant estimator is based partly upon the idea of particle filter, while the latter depends on the movement of each ant. For each ant estimator, the ldquopheromonerdquo update equation is well defined in order to guide ants to better solutions. Finally, Monte-Carlo runs are conducted and the results indicate that the two ant estimator perform well in estimating state parameters. In particular, we find that both are capable of tracking maneuvering target without any auxiliary means when employed in the target tracking field.