Evolutionary behavior learning for action-based environment modeling by a mobile robot

摘要: This paper describes an evolutionary way to acquire behaviors of a mobile robot for recognizing environments. We have proposed Action-based Environment Modeling (AEM) approach simple recognize In AEM, behavior-based acts in each environment and action sequences are obtained. The transformed into vectors characterizing the environments, identifies environments with similarity between vectors. suitable like wall-following AEM been designed by human. However design is very difficult him/her because search space huge intuitive understanding hard. Thus we apply robotics such using genetic algorithm make simulations which recognizes different structures. As results, find out learned even human hardly designs them, more efficient than hand-coded ones.