A study of the use of fuzzy control theory to stabilize the gait of biped robots

Abstract

SUMMARY This paper designs a biped robot to perform appropriate walking exercises according to the terrain, which then walks stably on a flat environment. The concept of fuzzy logic is combined with the Linear Quadratic Regulator (LQR) controller theory to design the best method to allow the biped robot system to have a balanced and stable gait. Traditional controllers are designed using mathematical models of physical systems, but a fuzzy controller is a physical system that uses an inexact mathematical model, which involves sets and membership functions. Fuzzy controllers use fuzzification, fuzzy control rules, and defuzzification. The method and theory of control : A stable gait for robots is achieved using inverse kinematics, fuzzy concepts, the LQR controller theory, path design, and the characteristics of a dynamic equation. It is then simulated using mathematical tools to prove that the system eliminates swinging by biped robots without fuzzy control knowing beforehand the dynamic model the system is using. Proportional-Integral-Differential control achieves a stable gait design in a flat environment.