A Dynamic Balancing Approach for a Quadruped Robot Supported by Diagonal Legs

Abstract

For legged robots, the most important task is to keep balance. This paper proposes a new balance control approach. To simplify the control complexity, first, LQR (linear quadratic regulator) control was used to obtain stable state feedback for the model. Then, the 6-DOF model was stabilized by dividing the whole robot into three separate parts. After that, VMC (virtual model control) was used to change the configuration of the joints. The simulation results showed that the proposed method allowed the quadruped robot to walk stably, even when certain types of disturbance were exerted on the models. In the simulation model, to mimic real conditions, noise was added to the sensors; the algorithm was then verified as still suitable for the quadruped robot.