Fuzzy Posture Control for Biped Walking Robot Based on Force Sensor for ZMP

Abstract

In these days, biped walking robots are concentrated on a service robot. Biped walking robots are suitable in a human life work such as walking up and down stairs and crossing a threshold because they have two leg mechanisms like a human. However, it is very difficult subject that robot maintains valance in walking. One way to solve this problem is using an accurate dynamic model of biped robot in 2-Dimension and 3-Dimension. Another method to solve this problem is modeling biped robot simply and controlling the robot with an adaptive control, intelligent control and robust control algorithm and so on. ZMP(zero moment point) concept was invented in this field. In this paper, ZMP trajectory following for biped walking robot based on force sensors and posture control using fuzzy algorithm are studied. The posture control of the biped robot is performed by ZMP information in static standing. Posture of biped walking robot is controlled by fuzzy algorithm referring to the obtained ZMP trajectory information. The result of this study is confirmed by experiment to apply fuzzy posture control against disturbance and to keep valance on a slope.