Single parent mating in genetic algorithm for real robotic system identification

<span lang="EN-US">System identification (SI) is a method of determining mathematical model for system given set input-output data. A representation made using based on certain specified assumptions. In SI, structure selection step where perceived as an adequate selected. typical rule that the final must have good balance between parsimony and accuracy. As popular search method, genetic algorithm (GA) used selecting structure. However, optimality depends much effectiveness GA operators. This paper presents mating technique named single parent (SPM) in use real robotic SI. chromosome parents such sufficient achieving eases optimal model. The results show three different objective functions (Akaike information criterion, Bayesian criterion parameter magnitude–based 2) respectively, with able to find more models than without technique. Validations selected are acceptable.</span>