Cartesian Genetic Programming Based Optimization and Prediction
This paper introduces a CGP (Cartesian Genetic Programming) based optimization and prediction techniques. In order to provide a superior search for optimization and a robust model for prediction, a nonlinear and symbolic regression method using CGP is suggested. CGP uses as genotype a linear string of integers that are mapped to a directed graph. Therefore, some evolved modules for regression polynomials in CGP network can be shared and reused among multiple outputs for prediction of neighborhood precipitation. To investigate the effectiveness of the proposed approach, experiments on gait generation for quadruped robots and prediction of heavy precipitation for local area of Korean Peninsular were executed.
KeywordsCartesian Genetic Programming gait optimization heavy rain prediction symbolic regression
Unable to display preview. Download preview PDF.
- 2.Miller, J.F., Jo, D., Vassilev, V.K.: Principles in the Evolutionary Design of Digital Circuits - Part I. Genetic Programming and Evolvable Machines 1, 8–35 (2000)Google Scholar
- 12.Korean Meteorological Society, Introduction to Atmospheric Science. Sigma Press (2009)Google Scholar