Development of monitoring algorithm for controlling a biped robot: norm bounded uncertainties system-based

Abstract

It is well known that the biped walking gait is represented as a steady periodic gait. Investigation of such passive natural motion leads to different strategies of control; these strategies require a sharp mathematical model of the walking dynamics. One of the most used models is the Kajita's one. In this paper, we present an algorithm of controller's design used for the stabilisation of biped robot's gait. Using the Kajita's model as a reference, we include a Norm bounded uncertainties to ensure a more realistic numerical model. The modified model allows us to include constraints on both inputs, outputs and states. The synthesis of the dynamic controller relies on the use of predictive control theory (MPC) and the resolution of a convex optimisation problem with linear matrix inequalities (LMIs) at every sampling period. The generated control law allows a real-time walking robot even in rough ground or unknown environment.