Robust Balance of Humanoid Robots Through a LQR with Integral Action

In this article, an approach to enhance the balance control in humanoid robots through a LQR controller with integral action is presented. The primary objective of this research is to significantly reduce the impact of disturbances generated during the robot’s interaction with its environment. To evaluate the effectiveness of this technique, its implementation is carried out on a NAO robot, followed by extensive experimental testing. These tests encompass both simulations using Webots software and implementations on a physical robot. Furthermore, an efficiency comparison between the proposed controller and a conventional LQR controller is made, highlighting the advantages of the proposed approach.