Abstract
Grammatical Evolution (GE), which is one of Evolutionary Computations (EC), is designed to find function or program satisfying the design objective. The candidate solutions are defined as the binary bit-strings. The genotype (bit-string) is translated into the phenotype such as function or program according to the translation rule which is defined in Backus Naur Form (BNF). The aim of this study is to apply GE to the design of LEGO Mindstorms EV3 control program. The designed program applied to the robot in real world after adjusting control parameters for bridging the gap between simulator and real world. As a result, the behavior in real world was roughly equivalent to in simulator.
