Big Bang - Big Crunch optimization algorithm hybridized with local directional moves and application to target motion analysis problem

Big Bang - Big Crunch (BB-BC) optimization algorithm relies on one of the theories of the evolution of the universe; namely, the Big Bang and Big Crunch Theory [1]. It was proposed as a novel optimization method in 2006 and is shown to be capable of quick convergence. In this work, local search moves are injected in between the original “banging” and “crunching” phases of the optimization algorithm. These phases preserve their structures; but the representative point (“best” or “fittest” point) attained after crunching phase of the iteration is modified with local directional moves using the previous representative points. This hybridization scheme smoothens the path going to optima and decreases the process time for reaching the global minima. The results over benchmark test functions have proven that BB-BC Algorithm enhanced with local directional moves has provided more accuracy with the same computation time or for the same number of function evaluations. As a real world case study, the newly proposed routine is applied in target motion analysis problem where the basic parameters defining the target motion is estimated through noise corrupted measurement data.