The Role of Fuzzy Logic Control in Evolutionary Robotics

Abstract

. This paper presents an evolutionary learning algorithm to facilitate the design of fuzzy controllers for mobile robots. It discusses the concepts, feasibility, benefits and limitations of current evolutionary techniques for fuzzy rule discovery and tuning. We propose an evolution strategy that optimizes the gain factors in the conclusion part of TakagiSugeno -Kang type fuzzy rules. We describe two applications of genetic-fuzzy systems in detail, adapting a wall-following behavior of a mobile robot and designing an autopilot for a small-size model helicopter. 1 Introduction Traditional AI approaches decompose robotic behaviors into a sense-model-planact type of hierarchy. The sensors provide perceptual information, which is used to build a model of the current environment. The planner generates a plan that enables the robot to accomplish the given task. A controller executes the actions commanded by the planner without taking novel sensor information into account. The utility of this...